Qualitative Research

Basics: A Guide for

Engineering Educators
Nancy Van Note Chism
Elliot Douglas
Wayne J. Hilson, Jr.

Rigorous Research in Engineering Education

NSF DUE-0341127, 2008

Introduction
It is our intention in this manual to provide an overview of the use of qualitative
research methods in the engineering education context. Our assumption is that
users of this guide will be fairly new to qualitative approachesand perhaps new
to educational research in general. We have tried, therefore, to avoid extensive
citations and detail, focusing rather on a general synthesis of the main issues
and approaches.
Because qualitative research rests on a different set of philosophical
assumptions about the nature of truth and limitations of research than traditional
research, the opening chapter begins with an explanation of these fundamental
differences. Throughout the following chapters, the ways in which these
assumptions play out in researcher choices are documented.
The chapters cover the essential information needed to conduct or evaluate a
qualitative study. They describe the design of qualitative studies, basic data
collection strategies, analysis techniques, trustworthiness considerations, and
writing up the study. Resources for additional information are listed in the final
chapter.
The power of qualitative research lies in its ability to adapt to natural settings
such as the classroom or laboratory; enable exploration of motivations,
reasoning, decision making, and other inner thoughts of participants such as
students and teachers; and permit description of the interaction of context and
actors in specific settings. For many research questions in engineering
education, it is the ideal research approach. It is relatively new in some
disciplines, however, and often regarded with suspicion. We hope that this guide
will help many researchers to see the utility of the approach, use it well, and
enrich the field with new understandings and informed practice.
Nancy Van Note Chism
Elliot Douglas
Wayne J. Hilson, Jr.

The Authors
Nancy Van Note Chism (general editor and author of Chapters 3, 5, 6 and 8) is
Professor of Higher Education and Student Affairs at the Indiana University
School of Education, Indianapolis. She has taught qualitative research methods
courses for over twenty years at Indiana University-Purdue University
Indianapolis and The Ohio State University. She has been involved in several
engineering education projects during her career, including the Gateway
Coalition and the Rigorous Research in Engineering Education projects, both
sponsored by the National Science Foundation.
Elliot Douglas (author of Chapters 1, 2, and 4) is Associate Professor of Materials
Science and Engineering at the University of Florida. He has been involved in
faculty development activities since 1998, through the ExCEEd Teaching
Workshops of ASCE, the Essential Teaching Seminars of ASME, and the NSFsponsored SUCCEED Coalition. He is a member of the American Chemical
Society, American Society for Engineering Education, and the American
Educational Research Association. His interests in engineering education are
focused on qualitative research methods, active learning techniques, problemsolving, and critical thinking. He is the Editor-in-Chief of Polymer Reviews and
has served as Chair of the Division of Polymeric Materials: Science and
Engineering of the American Chemical Society.
Wayne J. Hilson, Jr. (author of Chapter 7) is Director of the Teachers Resource
Center at the Indiana University School of Education, Indianapolis, and a doctoral
student in Higher Education and Student Affairs with a minor in Mathematics
Education. His B.S. and M.S. degrees were completed at Purdue University in
Mechanical Engineering and Education respectively. After working in industry
following the completion of his bachelors degree, he served as both Assistant
and Interim Director of Minority Engineering Programs at Purdue.

_______________________________________________________
This work was funded by National Science Foundation grant, DUE-0341127,
Rigorous Research in Engineering Education: Creating a Community of Practice.

ii

Contents
1.

Epistemological Perspectives

2.

Research Strategies

3.

Designing the Study

16

4.

Data Collection Methods

23

5.

Interviews

23

Observation

29

Documents and Materials

32

Maintaining and Judging Quality in Qualitative

Research

37

6.

Data Analysis

43

7.

Writing Up Qualitative Research

50

8.

Resources

61

iii

1. Epistemological Perspectives

What is Qualitative Research?

To the novice researcher, the terms qualitative and quantitative imply something
about the type of data that are collected. Thus, it may appear that the distinction between
these two research approaches lies solely in the data; qualitative research is concerned
with text or other non-numerical data, while quantitative research is concerned with
numerical data. This distinction, however, is only at the surface. The differences between
the two are much deeper, and go to the context in which the research is conducted.
In general, quantitative research is concerned with identifying relationships between
variables, and generalizing those results to the world at large. In contrast, qualitative
research seeks to understand phenomena in depth and within specific contexts. Denzin
and Lincoln (2005) state that, qualitative researchers study things in their natural
settings, attempting to make sense of, or to interpret, phenomena in terms of the
meanings people bring to them (p. 3), while Merriam (2002) says, Qualitative
researchers are interested in understanding what those interpretations (of those studied)
are at a particular point in time and in a particular context (p. 4). Thus, key
characteristics of qualitative research are an in-depth focus on only a few individuals or
situations, a focus on the context of the study, and recognition of the researcher as an
instrument of the study.
Qualitative research, in its approach, assumptions, and outcomes, is very different from
quantitative research. Questions of what constitutes truth, sample size and selection, how
data are collected and analyzed, how findings are presented, and what is meant by
trustworthiness and generalizability, are all different when one is conducting qualitative
research. Denzin and Lincoln (2005) discuss five differences between the two approaches
(pp. 11-12). According to them, qualitative research tends to have the following
attributes:
Less use of positivist or postpositivist perspectives
Acceptance of postmodern sensibilities
Capturing the individuals point of view
Examining the constraints of everyday life
Securing rich descriptions.
Thus, qualitative research serves as a complement to quantitative research. While the
general goal of quantitative research is to identify what is occurring in terms of contextfree generalizations, the goal of qualitative research is to gain an understanding of why or
how phenomena occur in terms of context-specific descriptions.

Epistemological Perspectives
The epistemological perspective of a study is related to the assumed nature of truth and
the approach used to understand and explain reality. Why is the epistemological
perspective of a study important? Because it helps to shape the research design and
makes explicit to others the phenomenon being investigated. Perhaps you are interested
in determining whether students with a higher SAT score do better in engineering. Or you
want to understand how the interactions between students in a problem-solving group do
or do not contribute to better problem solutions. Or you want to show how particular
faculty hiring practices tend to marginalize under-represented minorities. Each of these
studies is informed by a different view of reality, or epistemological perspective, which
in turn will shape the research design.
When designing a qualitative study, one of the first questions you must answer is what
epistemological perspective will drive the research. This perspective serves as an
umbrella to the entire research design, ensuring that there is consistency among research
questions, the research approach, the methods, and the data analysis. Table 1, taken from
Koro-Ljungberg and Douglas (2008), summarizes key aspects of the common
epistemological perspectives that can be used. Further details regarding each of these
perspectives appear in the following sections.
Positivist/postpositivist
Most engineers are familiar with the positivist and postpositivist perspectives, which
assume that there is an absolute truth. In the positivist perspective this truth can be found,
while in the postpositivist perspective one can only say that current data is consistent with
an assumed truth, because a counter-example could always be found (the concept of
falsifiability). In either perspective, however, there is ultimately some universal truth that
is independent of context, observer, or observed. In engineering research this approach
makes sense. After all, the strength of a certain material should not depend on who made
the measurement or where the material was made. Use of this perspective in social
science research is generally associated with quantitative methods and its concern with
validity, reliability, and generalizability.
Interpretivist
Much qualitative research is based on an interpretivist perspective, which states that truth
is contextual, depending on the situation, the people being observed, and even the person
doing the observation. This concept can be difficult to grasp, but can be illustrated with a
simple example. Consider a study that examines the relationship between SAT scores and
student grades in engineering courses. From a positivist perspective the correlation
between the two could be used to make predictive decisions about student performance in
engineering based on SAT scores. Thus, this correlation is assumed to represent a
generalized truth about the success of students. From an interpretivist perspective,
however, one would examine each students situation. Students who have similar SAT
scores and course grades might in fact be affected by different factors, such as inadequate

math preparation, test anxiety, etc. The qualitative study tells the individual stories of
these students and identifies the many truths that underlie the quantitative correlation
Thus, the quantitative, positivist approach focuses on context free generalizations, while
the qualitative, interpretivist approach focuses on context-dependent descriptions.
Under the broad umbrella of interpretivism, there are several epistemological
perspectives that provide additional focus to how reality is defined:
Phenomenology: Phenomenology seeks to provide a description of the key
elements of a phenomenon. New meaning can be obtained by laying bare the
essential aspects of a particular experience and describing what are the common,
key elements that make up that experience. Thus, in phenomenology researchers
ask for a description of the phenomenon that is readily apparent to conscious
minds. They do not attempt to describe how individuals arrived at a particular
meaning for the experience. Rather, they seek to obtain a careful description of
ordinary conscious experience of everyday lifea description of thingsas one
experiences them (Schwandt, 2001, p. 191).
Within phenomenology, there are two distinct schools. The classic European
approach is critical, in the sense that it recognizes there are culturally defined
meanings attached to our perceptions. The term bracketing is used to describe
the attempt to set aside the meanings brought by both the researcher and the
participants, in order to identify the true essence of the experience. Thus, its goal
is to identify the true essence of the experience, unfiltered by either the researcher
or the person experiencing the phenomenon. Critics of this approach to
phenomenology point out that it resembles positivism in its assumption that there
is an invariant aspect of the experience that is true for everyone. Phenomenology
is also used in an uncritical way, in that the experience is described in terms of the
perception of the person experiencing it. This approach is more interpretive in
nature, in that it allows for differing perceptions depending on the individual.
However, some authors lament this second approach, arguing that by being
uncritical, there is at best a failure to capture new or fuller meanings or a loss
of opportunities to renew the understandings that already possess us (Crotty,
2003, p. 85). Ultimately, a researchers choice of which phenomenological
approach to use depends on his or her own beliefs on the nature of experience.
Constructivism: Constructivism examines the meanings individuals create to
describe the world around them. Meaning occurs as a result of the individuals
interaction with the world and the particular biases of that individual. Thus, what
is known does not simply mirror the real world; rather, the meanings ascribed to
the real world are created by individuals as they experience it. A constructivist
viewpoint may seem clear for social concepts such as freedom. After all, a
particular individuals definition of freedom depends on his or her background,
culture, economic status, etc. But what of the natural world? Even here Crotty
(2003) argues for a constructivist (what he calls constructionist) stance (p. 57). He
describes different meanings for the word tree depending on our own personal

perspective: a source of income for loggers, a sink for carbon dioxide for
environmentalists, or an object of aesthetic beauty for an artist. A study based on
the constructivist perspective seeks to identify a particular meaning and how that
meaning was created by individuals.
Social constructionism: Similar to constructivism, and sometimes considered a
subset of it, this perspective describes meanings as being socially constructed, as
opposed to the individual construction of meaning in constructivism. the focus
here is not on the meaning-making activity of the individual mind but on the
collective generation of meaning as shaped by the conventions of language and
other social processes (Schwandt, quoted in Crotty, 2003, p. 58). Meaning thus is
created in a social context, through interactions between individuals. Meaning
reflects a shared sense of the world, which Crotty (2003) describes as, viewing
it (the world) through lenses bestowed on us by culture. Our culture brings things
into view for us and endows them with meaning, and, by the same token, leads us
to ignore other things (p. 54). Similar to the constructivist approach, then, a
social constructionist approach seeks to identify meaning and how that meaning
was created out of the social context that occurs between individuals. Social
constructionism is often considered critical, in that it seeks to examine how the
meanings are derived within a culture.
Symbolic interactionism: Symbolic interactionism focuses on how people act
towards the world around them, based on the meanings that are socially
constructed, but interpreted by the individual. This perspective focuses on
obtaining understanding by placing oneself in the place of the individual. Only
through dialogue can one become aware of the perceptions, feelings, and attitudes
of others and interpret their meanings and intent (Crotty, 2003, p. 75). The focus
is on understanding the interactions between people and the world, based on the
meanings given to the world by that individual. Thus, in contrast to social
constructionism, symbolic interactionism is very much uncritical. The researcher
accepts the meanings put into place by the culture and explores their implications.
Hermeneutics: Hermeneutics is primarily concerned with the analysis and
interpretation of texts. Interpretation proceeds through the hermeneutic circle,
which can be used to describe the relationship between the text as a whole and its
individual parts, between the object and the interpreter, and between the
interpreter and the interpreters background (Schwandt, 2001, p. 112). In the
hermeneutic circle, the individual parts of a text cannot be understood unless the
whole work is understood, but neither can one gain a meaning of the work as a
whole without understanding the parts. With regard to the interpreter, there is a
feedback in which the background brought to the text informs the interpretation;
however, that interpretation itself brings about new understanding in the mind of
the interpreter, which can change the way in which the text is seen. Thus, there is
an endless cycle of interpretation (the hermeneutic circle), which leads to endless
evolving interpretations of the text. There is no one true meaning for the text,

and new meanings can be discovered that even the original author did not
envision.
Critical
The critical perspective goes one step further than the interpretive perspective by
examining the power relationships present in society. It questions who has power, why,
and what the presence or absence of power means for particular groups. As such, it is
often focused on issues of race, class, or gender. As described by Merriam (2002),
Critical qualitative research, then, raises questions about the influence of race, class, and
gender (and their intersections), how power relations advance the interest of one group
while oppressing those of other groups, and the nature of truth and the construction of
knowledge (p. 10).
This last point by Merriam regarding the nature of truth is important. From the critical
perspective, truth is defined by those in power. Truth as defined in other ways is
marginalized, and those outside of the power structure are forced to either assimilate or
remain at the margins. Thus, the critical perspective is often associated with action
research, which seeks to not only understand, but to change the structures of society. By
giving a voice to those who are powerless, critical researchers seek to empower them and
expand the meanings ascribed to reality to include all views.
Postmodern/poststructural
The postmodern and poststructural perspectives question all aspects of the world,
especially grand narratives. These grand narratives are the theoretical frameworks that
describe social behavior. In place of grand narratives, the post perspectives emphasize
differences, multiple perspectives of the same phenomenon, and lack of absolutes.
Deconstruction is an important element of these perspectives. When data are
deconstructed, they are broken down to reveal hidden or latent meanings, revealing the
assumptions inherent in the data and their interpretation.

The Role of Epistemological Perspectives

The obvious question arises as to why epistemological perspectives are important at all.
After all, in engineering research this aspect of the research design is rarely considered.
In engineering research, it is well understood that the assumed perspective is positivist.
There is no need to explicitly state, or even consider, whether the researcher is using a
positivist perspective, because that is the only perspective that is used in engineering
research. Because there is a great diversity of possibilities when conducting qualitative
education research, however, it is important to consider the epistemological perspective
in research designs, in order to ensure that they are consistent with the assumed view of
reality and to make the epistemological perspective explicit in the reporting of the
research so that readers will understand the context in which the research was conducted.

Consider, for example, a study on how industrial design teams operate. The study can be
undertaken from a variety of perspectives:
From a phenomenological perspective, the interest would center on the
experience of being a part of a design team. The focus would be on describing
what happens in a team and generating a description that captures the essence of
how the team operates. The key question to ask would be: what are the essential
elements of the experience of working in a design team?
From a social constructionist perspective, the interest would be on knowing how
the design team creates a shared understanding of their activities, roles, or
understanding of the design process. The researcher would want to ask the
question: what are those understandings, and how did they develop?
From a critical perspective, the interest is with the power relationships in the
design team. The researcher would be asking questions regarding how those
power relationships were developed (formally or informally), how they influence
the working of the group, and what effect they have on those seen not to be in
power.
These examples illustrate that the epistemological perspective has a major influence on
defining the direction and scope of our research.

Summary
The epistemological perspective has an important influence on research. It provides an
umbrella for defining how the researcher sees the world and informs the research design.
Crotty (2003), in particular, proposes that the epistemological perspective determines
what research questions can be asked and what methods can be used. Certainly
consistency must be maintained among the epistemological perspective, research
questions, and methods for any particular study. This is not to say that any particular
researcher is limited to a single epistemological perspective. The design of a research
study is an iterative process, with the general purpose and research questions informing
the perspective, while at the same time the perspective informing the research questions.
Thus, you may find yourself conducting one study from a constructivist perspective, and
conducting another study that is critical. What is important is that any particular study is
consistent across the entire research design, so that the findings can be interpreted and
reported within the context that shaped the study.

References Cited in This Section

Crotty, M. (2003). The foundations of social research. Thousand Oaks, CA: Sage.
Denzin, N., & Lincoln, Y. (2005). Introduction: The discipline and practice of qualitative
research. In N. Denzin & Y. Lincoln (Eds.), The Sage handbook of qualitative
research (3rd ed.) (pp. 1-32). Thousand Oaks, CA: Sage.

Koro-Ljungberg, M., & Douglas, E. P. (in press). State of qualitative research in

engineering education: Meta-analysis of JEE articles, 2005-2006. Journal of
Engineering Education, 97.
Merriam, S. B. (2002). Qualitative research in practice: Examples for discussion and
analysis. San Francisco: Jossey-Bass.
Schwandt, T. A. (2001). Dictionary of qualitative inquiry (2nd ed.). Thousand Oaks, CA:
Sage.

Table 1: Comparison among epistemological perspectives. From Koro-Ljungberg &

Douglas (2008).
Epistemological
perspective

Post-positivist

View on reality

Single
falsifiable
reality

Purpose

The role of
researcher

To find
relationships
among
variables, to
define causeand effect
Methods and
variables
defined in
advance,
hypothesis
driven
Researcher is
detached

Outcome or
research
product

Context-free
generalizations

Methods

Critical/
emancipatory

Postmodern/poststructural

Multiple fragmented
realities

To describe a
situation,
experience, or
phenomenon

Multiple
subjective
and political
realities
To produce a
sociopolitical
critique

Methods and
approaches
emerge and are
to be adjusted
during study

Methods and
approaches
designed to
capture
inequities

Methods and approaches

generated during the
study

Researcher and
participants are
partners

Researcher
and
participants
are activists
Critical
essays,
policy
changes

Researchers and
participants have various
changing roles

Interpretivist
(constructivism,
social
constructionism,
hermeneutics,
phenomenology,
symbolic
interactionism)
Multiple
subjective
realities

Situated
descriptions

To deconstruct existing
grand narratives

Reconceptualized
descriptions of the
phenomenon

2. Research Strategies

The Role of Research Strategies

Many authors refer to specific qualitative research approaches, traditions, or strategies.
Denzin and Lincoln (2005) call them, a bundle of skills, assumptions, and practices that
the researcher employs as he or she moves from paradigm to the empirical world . . .
strategies of inquiry connect the researcher to specific methods of collecting and
analyzing empirical materials (p. 25). However, one should not consider these research
strategies as specific approaches that must be followed. After all, one of the inherent
strengths of qualitative research is its flexibility. As noted by Denzin and Lincoln, The
qualitative researcheruses the aesthetic and material tools of his or her craft, deploying
whatever strategies, methods, and empirical materials are at handIf the researcher
needs to invent, or piece together, new tools, he or she will do so (p. 4). In addition, they
note that, No specific method or practice can be privileged over any other (p. 7).
Within these limitations, descriptions of qualitative research strategies can be useful.
They provide a means to situate ones research within a larger body of research and to
design a study within the context of that particular strategy. They act as guidelines and
help to ensure that the study design is consistent with accepted qualitative research
practices.

Types of Research Strategies

The flexibility of qualitative research means that it can be hard to identify a particular
study with a particular approach. Denzin and Lincoln (2005) talk about blurred genres,
meaning that a particular study might have characteristics of more than one type of
approach. In categorizing the approaches, different authors have identified different sets
of strategies. Creswell (2007) identifies five main approaches:
narrative
phenomenology
grounded theory
ethnography
case study
Merriam (2002) identifies the following eight types of research:
basic interpretive
phenomenology
grounded theory
case study

ethnography
narrative analysis
critical research
postmodern research

Denzin and Lincoln (2005) list the following eight research strategies:
case study
ethnography/participant observation/performance ethnography
phenomenology/ethnomethodology
grounded theory
life history
historical method
action and applied research
clinical research
As these lists show, not all research strategies reflect the same aspect of the research
design. Phenomenology is a theoretical perspective that also has important elements of
data analysis, grounded theory is primarily a data analysis method, and case study refers
to the bounds of sampling. Merriam (2002) notes this, pointing out that case study
research can be approached as an ethnography or used to develop a grounded theory (p.
8). Six research strategies are highlighted here: ethnography, phenomenology, grounded
theory, case study, narrative analysis, and basic interpretive research.
Ethnography
An ethnography is a description and interpretation of the cultural behavior of a group. By
culture is meant the behaviors, language, and social practices that define the unique
aspects of that group. While one might typically think of a culture as referring to ethnicity
or religion, any group, such as an institution or organization, can be considered to have its
own culture. Observation of a culture can be used to understand the meaning of the
behavior that occurs within that group.
An ethnography involves in-depth immersion by the researcher in the culture being
studied. The researcher is often called a participant-observer in that the researcher
becomes, as much as possible, a member of the group while at the same time collecting
data. Data collection techniques involve interviews, collection of documents and other
artifacts, and observations. Ethnographers often record their observations in a field
notebook, where they can also record their personal thoughts and impressions. Analysis
of the data involves both creating a description of the groups behavior as well as analysis
of what that behavior means.
In some cases a study is called an ethnography if it involves any kind of participant
observation. However, to truly be an ethnography it needs to contain the six features
identified by Punch (1998):
The behavior of the group is understood by understanding the culturally shared
meanings within the group.
10

The study is conducted to understand meanings from the perspective of the people
in the group.
The group is studied in its natural setting.
The design evolves during the study, rather than being pre-defined.
A variety of data collection techniques are used within a single study.
Data collection occurs over a long period of time.

Thus, the focus of an ethnography is on understanding and interpreting the cultural

practices of a group through prolonged observation as a participant-observer.
Phenomenology
As an epistemological perspective, phenomenology seeks to identify the single, invariant
essence of the lived experience. Thus, phenomenology as a research strategy involves a
set of methods and analysis techniques that are used to identify that essence. Creswell
(2007) lists the following procedural steps involved in conducting a phenomenological
study (pp. 60-62):
An understanding of the philosophy of phenomenology, including the concept of
bracketing, is needed.
Research questions are developed that are focused on having the participants
describe their experiences.
Interviews are conducted with 5-25 participants who have experienced the
phenomenon being investigated. The questions asked focus on how they have
experienced the phenomenon and what contexts affect their experiences.
The data are analyzed, typically by first dividing the transcripts into specific
statements, clustering the statements based on psychological concepts, and then
bringing these clusters together to create a general description of the experience.
The final description of the experience provides an understanding of the
underlying, invariant structure that is common for all instances of the experience.
Since phenomenology is focused on the lived experience of the participants, the
researcher must bracket, or set aside, any pre-conceived notions of what that
experience is. This is not to say that they are simply ignored. Rather, as with qualitative
research in general, these pre-conceived notions are examined and made explicit in order
to be transparent about any biases the researcher may have. By making these biases
apparent, it is then possible to set them aside when examining the data. As described in
Chapter 1, however, this approach to phenomenology has been criticized. Specifically,
two beliefs tend toward a positivist perspective: the belief that there is a single true
essence of any experience and the belief that the researchers own interpretation can be
separated out in a way that will not influence the identification of that essence. This does
not mean that phenomenology is not useful, but as with any approach one needs to be
aware of its potential limitations.

11

Grounded Theory
Grounded theory is an approach to qualitative data that aims to develop theory that is
grounded in data. Grounded theory was first described by Glaser and Strauss in 1967,
and has since been refined and expanded. It was originally developed in reaction to the
trend within sociology to propose theories and then test them (Creswell, 2007). In
contrast, Glaser and Strauss emphasized the development of theory based on empirical
data. The approach is inductive rather than deductive.
Grounded theory is primarily a data analysis technique, and details of the data analysis
are covered in Chapter 6. Some of the key attributes of a grounded theory study are:
Data collection and analysis occur simultaneously. As analysis occurs, missing
elements are identified, which become targets for further data collection.
Data analysis proceeds through the constant comparative method, in which
interview statements are coded, categorized, and grouped to develop a theory.
Data collection proceeds until the categories are saturated, that is, until no new
information is obtained with further data collection.
The theory developed is not necessarily high level theory. It is often what
Creswell calls a substantive-level theory, which applies to the specific setting or
context in which the data was collected.
Some authors, Charmaz (2006) in particular, argue that classic grounded theory is
positivist in its theoretical perspective. This is seen in the original description of
grounded theory that implies that the data are discovered by the researcher and are
untouched by the competent researchers interpretations (Charmaz, 2006, p. 132). In
contrast, Charmaz advocates a more flexible view of grounded theory. In her view of
constructivist grounded theory, the researchers perspective becomes part of the data
analysis and interpretation. There are no objective data waiting to be discovered; rather
data and analysis are, created from shared experiences and relationships with
participants and other sources of data (p. 130). Further, the theory depends on the
researchers view; it does not and cannot stand outside of it (p. 130, italics in original).
Charmazs view moves grounded theory closer to the interpretivist perspective than
classic descriptions.
Case Study
A case study is characterized by in-depth study of a bounded system. The boundaries of
the case may be defined in terms of time, space, or participants (Merriam, 2002). For
example, a case may be a classroom, a particular event, or an organization. A case study
is selected when one wants to study a specific situation (the case) in depth. Within this
broad definition of a case study, there are many possible approaches that can be taken,
depending on the overall purpose of the study. For example, Stake (2005) identifies three
different types of case studies. The intrinsic case study has as its focus an in-depth
understanding of that particular case. The case is not chosen because it is representative
of a more general phenomenon, but simply because it is interesting in and of itself. The
instrumental case study is chosen to provide insight into a more general phenomenon.

12

The instrumental case may be chosen because it is representative of that phenomenon, or

because it is unusual and provides a useful contrast to the typical case. Finally, the
collective case study involves study of several cases that, taken together, provide insight
into the general phenomenon.
Punch (1998) lists four characteristics of a case study (p. 153):
The case has boundaries. Even if the boundaries are not obvious, the researcher
needs to define boundaries to limit the case.
The case is a particular example of something.
In analysis the researcher attempts to maintain the case as a single unit. At the
same time however, there is often a focus on particular aspects of the case.
A variety of data types and data collection methods are used.
Within a case study there is great flexibility on how to design the research. Data
collection methods can include interviews, observation, and collection of archival
materials. Analysis can be holistic, examining the entire case, or it can be focused on a
specific aspect of the case. Specific data analysis techniques vary, and can include
aspects of the strategies discussed above, such as phenomenology or grounded theory.
Case studies can also be conducted on single cases, or can be conducted across multiple
cases in order to obtain a broader view of the phenomenon.
In summary, Stake (2005) identifies the following responsibilities for researchers
conducting case studies:
Bounding the case, conceptualizing the object of study
Selecting phenomena, themes, or issues (i.e., the research questions to emphasize)
Seeking patterns of data to develop the issues
Triangulating key observations and bases of interpretation
Selecting alternative interpretations to pursue
Developing assertions or generalizations about the case.
Narrative
A narrative, or biography, describes a single persons experiences in order to identify and
understand larger meanings that make up an individuals story. A narrative is clearly very
individualized, but can also reveal broader truths that cut across many peoples
experiences. It is written as a story that leads the reader through the lived experiences of
the individual. There are many forms of narrative. Creswell (2007, pp. 54-55). Merriam
(2002) lists three approaches that a narrative can take (p. 287):
Biographical: The analysis is focused on significant events, people, and other
biographical elements of the persons life.
Psychological: The emphasis is on personal thoughts and motivations, and how
the persons experiences shape those thoughts and motivations.
Linguistic: The specific language used by the subject of the narrative is analyzed
using discourse analysis.
Creswell (2007) lists several steps in developing a narrative (pp. 55-57):
13

Gather the stories of the individual through multiple data sources.

Collect information on the context within which these stories occur.
Restory these experiences into a general framework that may link key elements,
provide a chronological story, or explore larger themes.
Explore the meaning of these stories, with input from both the researcher and the
subject.

Although narrative analysis is focused on the experience of single individuals, when

framed properly it can provide insights into larger issues that cut across multiple
experiences. For example, Merriam (2002) reprints a narrative study in which a woman
in a management position in a large corporation confronted and dealt with issues of
sexism (Bloom, 1996). Use of narrative can be a powerful tool for illuminating and
understanding broader issues beyond the single individual.
Basic Interpretive
Merriam (2002) defines a basic interpretive study as one in which the overall purpose is
to understand how people make sense of their lives and their experiences (p. 38) (italics
in original). While this goal underlies all qualitative studies, Merriam states that studies
based on other strategies have additional purposes not found in basic interpretive studies
(e.g. developing theory in grounded theory, identifying the essence of a phenomenon in
phenomenology, etc.). For Merriam, the basic studies are those that do not share the
special characteristics of the other strategies. She gives the example of a study that
examines adult learning in a non-Western culture. Although the culture was an important
aspect of the interpretation, this is not an ethnography because it did not involve a
prolonged stay or the researcher as a participant-observer.

Summary
Research strategies are a set of qualitative research practices that have their own methods,
history, and literature. Identifying which research strategy you are using in your own
study can be useful to help situate your study in a particular literature and to guide your
choice of data collection and analysis methods. At the same time, however, the types of
research strategies enumerated in the literature should not be considered prescriptive.
Qualitative researchers need to maintain their flexibility, selecting methods that answer
their research questions and are consistent with the theoretical perspective of the study.
Merriams identification of basic interpretive studies as a research strategy highlights this
flexibility. As she points out, They are probably the most common form of qualitative
research found in education Careful consideration of whether or not your study falls
into one of these strategies, and if so, which one, can provide guidance as you move into
the research design.

References Cited in This Section

Bloom, L. R. (1996). Stories of one's own: Nonunitary subjectivity in narrative
representation. Qualitative Inquiry, 2(2), 176-197.
14

Charmaz, K. (2006). Constructing grounded theory. Thousand Oaks, CA: Sage.

Creswell, J. W. (2007). Qualitative inquiry and research design: Choosing among five
approaches (2nd ed.). Thousand Oaks, CA: Sage.
Denzin, N. K., & Lincoln, Y. S. (2005). Introduction: The discipline and practice of
qualitative research. In N. K. Denzin & Y. S. Lincoln (Eds.), The Sage handbook
of qualitative research (3rd ed.). (pp. 1-32). Thousand Oaks, CA: Sage..
Merriam, S. B. (2002). Qualitative research in practice: Examples for discussion and
analysis. San Francisco: Jossey-Bass.
Punch, K. R. (1998). Introduction to social research: Quantitative and qualitative
approaches. Thousand Oaks, CA: Sage.
Stake, R. E. (2005). Qualitative case studies. In N. K. Denzin & Y. S. Lincoln (Eds.), The
Sage handbook of qualitative research (3rd ed.). (pp. 443-466). Thousand Oaks,
CA: Sage.

15

3. Designing the Study

Designing a qualitative study is similar in many respects to other types of studies, yet
different in some key ways. Like quantitative studies, one must identify the research
questions of the study, align the data collection methods with those research questions,
and consider the appropriate population to study. However, qualitative studies have
additional considerations, such as the epistemological perspective that underlies the study
and the role of the researcher in both data collection and analysis. This section will
describe the major characteristics of qualitative design and elements that should be in a
study proposal, such as decisions about data collection, sampling, Human Subjects, and
ethics.

Characteristics of Qualitative Studies

As Chapter 1 described, in accord with the epistemological assumptions on which they
rest, qualitative studies have the following design characteristics:
Naturalistic orientation: The studies are almost always conducted in the natural
setting, rather than under laboratory conditions. They do not normally rely on
interventions or manipulations, but seek to understand existing conditions.
Purpose: The studies are not primarily aimed at prediction, control, or
generalization, but rather understanding, and in the case of critical studies,
critique.
Researcher involvement: In qualitative studies, the researcher is instrument,
an essential lens through which the reality of the participants is documented. The
researchers presence is acknowledged and his or her engagement is regarded as
an asset rather than a contaminating factor.
Holistic perspective: Rather than isolating variables, the researcher is interested
in looking at things in context and noting the patterns of interaction and change.
Unique case orientation: Depth is more important than breadth; individual
characteristics are the focus rather than attempts to find the typical or to make
statistical generalizations.
Inductive approach: The researcher works from particulars to patterns, not from
theories to specific cases. The approach is exploratory rather than confirmatory.
Theory is generated rather than tested; explicit hypotheses to be tested are not
formally developed before the study begins.

16

Emergent design: Rather than laying out a fixed design in advance, qualitative
researchers develop tentative plans and make adjustments as findings suggest new
directions. A study that began with interviews as the primary method of data
collection might turn to observation as well; one that anticipated being situated at
only one site might move to others, all based on decisions made in process.

Design Elements of a Study Proposal

Outlines of qualitative studies contain similar elements to quantitative studies, but these
are approached somewhat differently in line with the assumptions above. The following
list describes the major elements of a research outline from a qualitative perspective.
Janesick (2000), Marshall and Rossman (2002), and Schram (2006) all offer helpful
descriptions of study design.

Research topic and questions: The foundations of good qualitative studies, like
all other kinds of research studies, are good topics and questions. Topics should
lend themselves to empirical exploration rather than be matters of preference (Is
red better than blue?) or logic or metaphysics (Does God exist?). They should be
highly relevant to the interest of the researcher and the field. Research questions
should focus on a specific piece of the topic that is feasible for study. The
question, How is engineering best taught? is too broad and ill-defined. A better
question is, How do highly independent students navigate the experience of
working in cooperative learning groups in a freshman engineering course?
The major research question can usually be broken into several other subquestions. For the question above, sub-questions might include, How does
working in groups affect the students understanding of the subject matter? How
does it influence their social networking? How does working in groups advance or
slow down their preferred pace? How do the products that are generated in groups
differ in quality and style from those generated individually?

Significance: As with other studies, it is important to specify why the topic is

important to explore and of what use the findings will be.

Epistemological perspective: Qualitative studies often begin with a brief

explanation for why the epistemological assumptions of qualitative research fit
the purpose of the study. For example, it might be argued that to understand the
many ways in which first generation students experience being an engineering
major, it is important to adapt a constructivist approach that values the multiple
realities of these students. Some argue that since qualitative research has been in
use for some time, it is no longer necessary to justify its use (after all, how many
quantitative studies begin with an argument for the quantitative approach?).
However, due to the relative novelty of the approach in engineering education,
some discussion of the epistemological perspective during the design stage and in
the studys introduction still seems to be warranted.

17

Conceptualization: Given the inductive nature of qualitative studies, the

conceptual framework is viewed more tentatively than it is in traditional studies.
The emphasis is on discovery and understanding, so energy devoted in advance to
specifically describing the topic of interest, defining variables, and modeling it in
terms of established theories is regarded as potentially hampering the researcher
from seeing it in a new light. At the same time, it is important for the researcher to
articulate his or her assumptions about the topic and the ways in which it is being
approached. For example, thinking of incoming students as underprepared is
not the same of thinking of them as differently prepared. Understanding
teaching as information transmission is different from seeing it as facilitating
student engagement in learning. When the researcher spends some time
articulating how he or she thinks about the topic and its context, the questions and
methods of the study are more intentional.

Choice of tradition: As described in Chapter 2, there are a variety of approaches

and traditions that qualitative researchers use. Deciding whether the research
questions lend themselves to a phenomenological approach, an ethnography, a
grounded theory study, or some other form is the next design task.

Literature review: As with the conceptualization, there are some qualitative

researchers who argue that working from previous scholarship might prevent the
researcher from looking at the phenomenon of interest with fresh eyes.
Approaches, arguments, and findings of other researchers can reinforce the takenfor-granted. However, most scholars believe in the importance of building on
each others work and recommend that qualitative researchers explore the major
types of studies that have been conducted in the area in which they are doing
research. Unlike other researchers, however, qualitative researchers often broaden
their literature review while they are gathering data and during analysis as new
ideas and connections emerge. As with other literature reviews, this activity is
situated as an argument: what has been studied that is pertinent to my topic and
why is my study still needed?

Data collection strategies: To describe how the research topic will be explored,
the qualitative researcher next has to detail methods for data collection and
sampling.
o Methods: These are described in Chapter 4 of this handbook. It is important to
delineate in a research plan the answers to some basic questions for each type
of method. For example,
o If observation:
Entry (How will I gain access to the site and gain trust?)
Stance (Will I be a participant, observer, or both?)
Way of recording (Will I use audiotape, videotape, notes, all?)
o If interviewing:

18

 Interview type (Will I use an informal, structured, or semi-structured

approach?)
Way of recording (Will I use audiotape, videotape, notes, all?)
o If document or materials analysis
Obtaining (How will I gain access to these sources?)
Way of recording (Will I take notes, file actual documents?)
o Sampling: In addition to type of method, the researcher has to anticipate what
sites, participants, or artifacts will be explored, although in keeping with the
emergent nature of qualitative designs, these decisions may be modified
during the study. Sampling involves choosing both type and number/duration.
A key characteristic of qualitative research is its reliance on the purposeful
sample, rather than the random sample of traditional research. Instead of
relying on the law of large numbers and chance to select a sample, qualitative
researchers try to locate specific people or artifacts that are potentially the
most information-rich for their study. Based on their topic, their initial
assumptions, and the practicalities of their study, they develop an intentional
rationale for choosing their sample. Patton (2002) enumerates some of the
types of samples that a qualitative researcher might select in a convenient
chart (pp. 243-244). Some of the more commonly used approaches include:

Extreme or deviant case sampling: Looking at unusual instances to

arrive at an understanding of the essential characteristics and boundaries
of the topic. An example would be looking at entrants to engineering
programs who are extremely gifted in math or who have math anxiety.
Maximum variation sampling: Exploring as many different kinds of
cases along the continuum of experience with or knowledge of the topic,
such as including students from different subdisciplines, different years of
study, and different genders and ethnicities. In maximum variation
sampling, it is important to articulate why one expects to find differences
on any given dimension as the rationale for varying the sample on that
basis.
Snowball sampling: Seeking the assistance of initial study participants to
generate contacts for additional cases that are potentially highly
informative for the topic.
Criterion sampling: Selecting for study those people, places, or things
that meet certain specifications, such as female second-year students in
problem-based civil engineering courses. As with maximum variation
sampling, it is important to argue why adhering to a given criterion is
expected to produce a sample that is suited to the best exploration of the
topic.
Convenience sampling: Working with people or things that are easy to
access, such as ones own students, friends, or public Internet sources.
Although this approach might not yield the most information-rich sources,
it is probably the most frequently employed approach to sampling.

19

Determining the size of the sample also relies on intentional researcher choice
as well. Because of both the desire to explore in depth and the labor intensity
of data collection methods, qualitative researchers focus on small instances of
the topic they are exploring. Case studies frequently study only one person or
place or event. Often, researchers identify a tentative number of participants,
documents, or sites they plan to study and as the study unfolds, add or subtract
to this number based on the patterns they are seeing. Qualitative researchers
are advised to sample until redundancy; that is, until significant new
information or details are not appearing and the data are saturated. This
requires judgment and will vary according to the purpose and type of study.
Decisions about time in the field are sampling decisions that are as important
as decisions on participant selection. They should be made on the same basis
as other sampling decisionsthe researcher should remain on site until data
categories are saturated.

Mechanisms for trustworthiness: In this section of a study plan, qualitative

researchers anticipate what steps they will take to ensure that they have conducted
the study according to criteria for quality research. Particular strategies that may
be employed are described in Chapter 5.

Treatment of ethics: This part of the outline describes how the researcher will
ensure that his or her conduct as a researcher respects the site, the individuals,
and/or the artifacts involved in the study. It also entails other ethical
considerations, such as the responsibility to report data honestly. In the case of
conduct during the study, research with human subjects is governed by federal
laws which are designed to protect research participants from harm. In order to
comply with these laws, all institutions have an Institutional Review Board (IRB)
that reviews research plans. While details vary from institution to institution,
some general guidelines are as follows:
o All research, even studies that use standardized tests, must be reviewed by
the IRB. In some cases, particularly education studies, the research may be
considered exempt, meaning the IRB simply needs to be informed of the
study. Other studies may be eligible for expedited review, which means
that the full IRB does not need to review it, but a designee or smaller
group will do the review. Full review by the board is rarely required for
educational studies, but in cases where minors or sensitive information are
involved, it is a possibility.
o All studies require informed consent by the participants. Generally this
means that you will provide a form for the participants that describes the
study, what will be done, what benefits they will receive, and what the
potential harms are. At some institutions this form must have an official
approval stamp from the IRB before it can be used. This form must be
signed by the participants before data collection begins. The signed copies
must kept on file for potential audits. It is also good practice to provide a
second copy to the participants for their records.

20

o If there are changes to the research protocol during the study (e.g. there is
a need to increase the number of participants beyond what was approved
by the IRB), permission is generally needed from the IRB before the
change is made.
o Because of these requirements, it is important to begin the IRB approval
process well before the start of data collection. Depending on the level of
review, it can take several months for the study to be reviewed, and it is
not unusual for the IRB to request changes to your protocol or informed
consent form before granting approval. This issue is particularly
problematic with qualitative studies. IRBs were established largely to deal
with clinical medical studies, and therefore tend to be more familiar with
quantitative studies in which procedures are well-established before the
study begins. As has been discussed throughout this handbook, however,
one of the characteristics of qualitative studies is that the design emerges
throughout the study. There are no definitive ways to address IRB
requirements for qualitative studies, since it depends very much on the
local IRB. Researchers planning on conducting a qualitative study should
contact their IRB early in the process to find out requirements for getting
the study approved.

Method of analysis: The study plan contains stated expectations for how data
will be analyzed. Various methods for analyzing data are described in Chapter 6
of this handbook. The anticipated procedures for use in the study are outlined in
this section of a study plan.

Method of reporting: The researcher next anticipates what format will be used to
convey the studys findings and where the study may be reported. Ideas for this
part of the study plan are contained in Chapter 7 on writing the study.

Logistics and timeline: Finally, a study plan lays out a projected task list and
timeline. If multiple researchers are involved, specific responsibilities of each are
delineated. As with other aspects of the study plan, this may be modified if new
insights or logistical problems surface.

In addition to outlining a clear yet flexible study plan, designing a qualitative study
includes the activities of pilot testing interview protocols or other tools as well as
obtaining access to people, sites, or artifacts. Devoting time to careful and thorough
preparation during the design phase of a study will reap important rewards later.

References Cited in This Section

Janesick, V. J. (2000). The choreography of qualitative research design: Minuets,
improvisation, and crystallization. In N. K. Denzin & Y. S. Lincoln (Eds.),
Handbook of qualitative research (2nd ed.). Thousand Oaks, CA: Sage.

21

Marshall, C., & Rossman, G. B. (2006). Designing qualitative research (4th ed.).
Thousand Oaks, CA: Sage.
Patton, M. Q. (2002). Qualitative research and evaluation methods (3rd ed.). Thousand
Oaks, CA: Sage.
Schram, T. H. (2006). Conceptualizing and proposing qualitative research (2nd ed.).
Upper Saddle River, NJ: Pearson.

22

4. Data Collection Methods

Data collection is not simply the mechanical gathering of data for later analysis. As with
all aspects of the research design, data collection must be tied to the research questions
and epistemological perspective of the study. Crotty (2003) points out, Justification of
our choice and particular use of methodology and methods is something that reaches into
the assumptions about reality that we bring to our work. To ask about these assumptions
is to ask about our theoretical [epistemological] perspective (p .2). Table 2 is from KoroLjungberg and Douglas (2008), which links data collection methods to epistemological
perspectives. While one should not feel constrained by the links shown in this table, it
does emphasize that there is a connection between methods and perspective that needs to
be considered as part of the study design.
This chapter focuses on the most commonly used qualitative data collection methods of
interviews, observation, and documents and materials. The discussion covers not only
some of the practical aspects of data collection, but also how the different kinds of data
can be used and what assumptions are made in understanding the data.

Interviews
Interviews are a common method for collecting data in qualitative research. Interviews
allow the researcher to collect data from specific participants, focus the participants
attention on specific items of interest, and obtain the participants views on the topic.
From an interview the researcher learns about the experiences, perceptions, and feelings
of those being interviewed. However, it is important to remember that interviewing does
not produce an authentic re-creation of those experiences. Rather, interview data are
filtered through the perceptions of the interviewees and reflect their own personal views.
Thus, interviews are most appropriate when the researcher is interested in those views,
such as in phenomenological or constructivist studies. The interview is a complex process
that has many different elements and can be used in many different ways. In preparing for
an interview, one needs to consider the type of interview that is most appropriate to the
chosen research design, what types of questions to ask, and how to conduct the interview.
Each of these areas will be considered in turn, and the treatment of the interview will then
conclude with a discussion of the role of the interviewer in the process.
Types of Interviews
There are many different ways to classify interviews, based on the types of questions
being asked, the group that is being interviewed, or the purposes for which the interview
data are to be used. Interviews are generally classified into three types:

23

Structured: In this type of interview the interview questions are defined in

advance. The answers are generally intended to be short and fit into predetermined categories. This type of interview allows for standardization across all
interviews, but does not allow unexpected information to be obtained.
Semi-structured: The most common type of qualitative interview, it involves
pre-determined, open-ended questions, but also allows the interviewer to explore
themes that emerge during the interview process. Thus, the interviewer can ensure
that certain key elements are covered, but allow flexibility.
Unstructured: Unstructured interviews provide the most flexibility, but are also
the most difficult to conduct. Unstructured interviews can be either formal or
informal. The formal interview occurs in a setting in which there is a clear
interviewer/interviewee relationship. The interviewer only has a few, grand tour
questions prepared in advance to guide the interview process. The interviewer acts
primarily as a guide, allowing the participant to explore the areas defined by these
questions. The informal interview typically occurs in ethnographic studies in the
natural setting (e.g. a street corner, a bar, etc.). The interviewer may have general
topics to explore, but there are no pre-determined questions or categories.

Each of these types of interviews has its own purpose. As explained by Fontana and Frey
(2005), the difference between structured and unstructured interviews is, The former
aims at capturing precise data of a codable nature so as to explain behavior within
preestablished categories, whereas the latter attempts to understand the complex behavior
of members of society without imposing any a priori categorization that may limit the
field of inquiry (p. 706). In reality, these types of interviews only represent specific
points on a continuum, and the level of structure in an interview depends on the needs of
the particular study.
Another way of categorizing interviews is by whom is being interviewed, an individual or
a group. The group interview, or focus group, can be a powerful tool for gaining
information not accessible in individual interviews. Quite often interactions between
group members stimulate memories and ideas that would not have surfaced in an
individual interview (Fontana & Frey, 2005). Group interviews can be especially useful
for making apparent the views, perceptions, and motivations of the participants (Punch,
1998). However, there are some pitfalls inherent in the group interview. These include:
some members of the group may dominate the conversation; some members may be
reluctant to participate in the conversation; and there is the potential for groupthink to
occur, which limits the information that is obtained (Fontana & Frey, 2005). From the
interviewers standpoint, the group interview is more difficult to conduct than the
individual interview and requires more practice and experience.
Interview Questions
In preparing for an interview, it is important to develop a set of interview questions (often
called an interview guide or protocol) that is consistent with the type of interview you
intend to conduct and that will result in data that will answer your research questions.
There are several types of questions that have varying levels of specificity. Grand tour

24

questions are broad questions that invite an open-ended response. Focusing questions ask
for more specific responses to a broad question. Probes are used with focusing questions
to prompt responses when the interviewee has some difficulty responding or when the
interviewer seeks elaboration. During the interview itself you should generally start with
the broader questions, and work towards more specific questions as needed, particularly
if the person you are interviewing tends to give only short, not very descriptive answers
to the broad questions. Examples of each of these types of questions are:
Grand tour question: Tell me about your experience teaching design courses.
Focusing question: How do you choose design projects for your course?
Question with probes: How are students grouped into design teams in your
course? Do you assign the groups based on grades? Randomly? Do the students
select their own team?
Kvale (1996) further identifies interview questions that may be part of the original
protocol, or may be used to elicit additional information in the course of the interview:
Introducing questions: These are initial questions used to have the participant start
to describe the topic of the interview.
Follow-up questions: These are used to encourage the participant to expand on
points given in response to the introducing questions
Probing questions: These are general questions that seek to elicit additional
information
Specifying questions: The counterpoint to probing questions, these questions are
still open-ended, but ask for more specific information regarding statements the
participants have made.
Direct questions: These questions allow the interviewer to ask about specific
topics that the participant may not have brought up spontaneously.
Indirect questions: These are questions that ask the participant to project beyond
their immediate experience. These can be used to indirectly obtain information
about the participants own beliefs.
Structuring questions: Statements or questions that allow the interviewer to
maintain control over the interview. They are often used to end discussion on one
topic and begin a new one.
Silence: Silence on the part of the interviewer can often prompt spontaneous
statements from the participant that expand on previous statements.
Interpreting questions: These ask for interpretation on the part of the participant.
They encourage the participant to go beyond mere statement of fact and explore
the meanings of their experiences.
In developing questions, there are a number of common mistakes to avoid. First, it is
important to ensure that the questions are open-ended and will not result in a limited
response. For example, the question Does your professor give examples of how to solve
problems? will only result in a yes or no answer. A more appropriate question
would be How does your professor demonstrate problem-solving? This question will
elicit a more complete response, and still leaves open the possibility for the interviewee
to say He doesnt. Another type of question to avoid is one that asks several different
things. An example is the question, How does your professor demonstrate problem
25

solving, stress critical thinking, and use technology? It is better to phrase these into
separate questions. Internal bias is also something to avoid. The question How does
being the only woman on your design team cause conflict? assumes that the conflict is
caused by there being only one woman. A more appropriate question is What factors do
you see as causing conflict on your design team? This question allows the participant to
fully articulate her own view of what causes conflict.
Another important aspect of developing interview questions is ensuring consistency with
your epistemological perspective and overall research design. Consider as an example a
study of critical thinking of engineering students. If one were conducting this study from
a constructivist perspective, one interview question might be:
What kinds of engineering problems require critical thinking?
This question goes to the constructivist aspect of the study, in that it probes the students
own view of what is meant by critical thinking and how it can be applied in engineering.
The following question, although important, would not be consistent with a constructivist
perspective:
How do your engineering professors teach or demonstrate critical thinking?
Asking the student about how they have learned critical thinking from their professors
moves beyond the constructivist perspective towards social constructionism. Keeping the
interview protocol focused within the areas of the general research design ensures that the
specific research questions will be answered. In this example, while asking about the
professors may seem important, including that question in the protocol can lead to a study
that is unfocused and does not provide the data needed to answer the research questions.
Conducting the Interview
Preparing for an interview takes considerable planning. Punch (1998) lists several issues
to consider (p. 180):
Who will be interviewed and why?
How many will be interviewed, and how many times will each person be
interviewed?
When and for how long will each respondent be interviewed?
Where will each respondent be interviewed?
How will access to the interview process be organized?
Each of these aspects represents an important part of the research design process, and
depends on the epistemological perspective and research questions of the study. A
phenomenological study may involve single, one-hour interviews with 5-20 participants,
while a narrative study might involve many hours of multiple interviews with a single
person.

26

The interview itself is a complex process that requires considerable skill on the part of the
interviewer. Kvale (1996) lists ten characteristics of effective interviewers:
Knowledgeable
Structuring
Clear
Gentle
Sensitive
Open
Steering
Critical
Remembering
Interpreting
Depending on the type of interview being conducted, the interviewer must be flexible,
able to read the interviewee and guide the interview in directions that will elicit the
desired information. Fontana and Frey (2005) point out the importance of nonverbal
communication, such as body movement or changes in volume or pitch, in
communicating information by both the interviewer and interviewee. These nonverbal
aspects can provide important clues to the interviewer regarding the participants
thoughts and feelings on the subject being discussed.
From the practical standpoint, interview data can be recorded in numerous ways. Quite
often written notes are taken, especially in ethnographic informal interviews where the
use of a recording device would be intrusive. Fontana and Frey relate examples of
ethnographic studies in which the researcher would rush to a bathroom immediately after
a discussion and record notes on toilet paper (p. 708)! More common is the researchers
field notebook, in which notes can be taken. In more formal settings in which an
individual or group is brought to a specific location for an interview, audio- or
videotaping is common. Videotaping can be particularly useful for capturing the
nonverbal aspects of the interview. For structured interviews, a simple checklist where
the responses can be recorded may suffice.
After data collection, recorded interviews must be transcribed. In some cases it may not
be necessary to do a complete transcription; paraphrasing or transcription of only selected
portions will be sufficient. For the majority of studies, however, the entire interview
should be transcribed. When doing transcription, frequent rewinding of the tape to check
the accuracy of your transcription is recommended. Codes can be developed for
nonverbal aspects of the interview (e.g. pauses, tone of voice, etc.) if they are considered
to be important. As a general rule of thumb, one hour of tape will take three to four hours
to transcribe, although that will vary depending on the complexity of the data and the
experience of the transcriber. Digital recorders enable to transcriber to have access to the
file on the same desktop as the transcription file, and can be more conveniently
manipulated in terms of rewinding or stopping and starting. Although some experts
discourage researchers from hiring transcribers, fearing that they will lose an opportunity
to digest the data more deeply, practically speaking, it is often wise to have a professional

27

transcribe the recorded interviews, and to read these over carefully afterward against the
voice recording, revising rather than doing the entire transcription yourself.
The quality of the interview data has a significant impact on the overall quality and
trustworthiness of the outcomes of the study. Kvale (1996) lists the following criteria for
determining the quality of an interview:
The extent of spontaneous, rich, specific, and relevant answers from the
interviewee
The degree to which the interviewers questions are shorter than the subjects
answers
The degree to which the interviewer follows up and clarifies the meanings of the
relevant aspects of the answers
The extent to which the interview is interpreted throughout the interview
The frequency with which the interviewer attempts to verify his or her
interpretations of the subjects answers in the course of the interview
The degree to which the interview is self-containing it is a story contained in
itself that hardly requires many extra descriptions and explanations.
Recent Perspectives on Interviewing
The classic view of interviewing has tended towards a positivist perspective, seeking to
maintain the interviewer as a neutral party who simply records the information and has no
influence on the resulting data. More recently, however, it has been recognized that the
interviewer plays an important role, even when there is an attempt to maintain this
neutrality. Fontana and Frey (2005) discuss the empathetic view of interviewing,
pointing out that, interviewing is not merely the neutral exchange of asking questions
and getting answers. Two (or more) people are involved in this process, and their
exchanges lead to the creation of a collaborative effort called the interview. The key here
is the active nature of this processthat leads to a contextually bound and mutually
created story the interview (p. 696, italics in original).
How does this contextualization occur? One way is the how the interview is framed. How
one describes those being interviewed (pathological deviants vs. diverse individuals who
are ostracized) leads to different understandings and uses of the data (Fontana & Frey,
2005). Fontana and Frey also point out that the researcher influences the data by deciding
which aspects to include and which to exclude from the narrative that results from the
data. In considering the role of the researcher, there are a variety of approaches. At one
extreme, the role of the researcher is ignored, even though it is always present. At the
other extreme, steps are taken to enhance the interaction between interviewer and
interviewee, making them equal partners in creating the narrative. Researchers are
attempting to minimize status differences and are doing away with the traditional
hierarchical situation in interviewing. Interviewers can show their human side and can
answer questions and express feelings (Fontana & Frey, 2005, p. 711). By becoming a
partner in the interview, researchers can more explicitly recognize the role that the
interviewer plays in the interview and encourage those being interviewed to be more
open about their personal responses and emotions regarding the topic. Feminist scholars

28

are particularly adamant about power issues between researcher and participants, arguing
for reciprocity in interviewing.

Observation
Observation grew out of sociology and ethnography, initially focused on the study of
non-Western cultures. In contrast to interviews, observation has the potential to provide a
view of a particular setting that is unfiltered by those being observed. However,
observation is not always a straightforward task. There are many different ways to
conduct an observation, and the subjectivity of the researcher still needs to be considered
when interpreting the results of an observational study. Each of these issues is discussed
below.
Types of Observation
Punch (1998) distinguishes between direct observation and participant observation.
In his classification, direct observation occurs when the observer is an outsider. As he
puts it, observers neither manipulate nor stimulate the behaviour [sic] of those whom
they are observingThe situation being observed is not contrived for research purposes
(p. 185). In contrast, participant observation occurs in ethnographic studies, when the
observer is not neutral, but rather acts as a participant-observer. While this classification
is useful, it does not include observation in a non-natural setting as might be more
common in educational studies. For example, studies of problem-solving or design can be
conducted in a research setting where the participants are given a problem to solve that is
specifically created for the purpose of the research study.
Angrosino (2005) describes a similar set of observation categories (p. 732):
Participant observation: The researcher becomes a part of the community or group
being studied.
Reactive observation: Observation is conducted in a controlled setting, with the
participants aware that they are being observed.
Unobtrusive observation: The participants are not aware that they are being
observed.
Another classification scheme is whether the observation is structured or unstructured. In
structured observation there are specific categories of behavior that are identified ahead
of time. The observation protocol then involves identifying whether or not these
behaviors occur. In contrast, an unstructured observation occurs by recording in some
fashion everything that occurs. Subsequent analysis of the data is used to create meaning
from what occurred.
In reality, all of these categorizations represent points on a continuum of possible
observation protocols. Patton (2002) provides a list of six dimensions of observation.
Each of these dimensions is a continuum, ranging between the two extremes listed (p.
277):
Role of the observer: full participant to spectator

29

Perspective: insider to outsider

Who conducts the inquiry: professionals to people in the setting being studied
Disclosure of the observers role: full disclosure to no disclosure
Duration of observations: short, single observation to long-term, multiple
observations
Focus: single element to holistic view of the setting

Conducting an Observational Study

Prior to conducting an observational study, a number of decisions need to be made. As
with any study, the question of whom and what will be observed is paramount, and must
be consistent with the epistemological perspective and research questions. If one is
approaching the research from a constructivist perspective, observations of groups would
not be appropriate.
For a structured study, the specific data to be collected must be identified ahead of time.
In an unstructured study, the process evolves as data collection proceeds. This process
has been described as a funnel (Punch, 1998). Initially there are only broad questions
that the researcher wants to answer. As observation proceeds and data are collected, the
research questions are refined in light of the data collected, and the focus of the
observations becomes tighter in order to narrow in on these revised questions. Punch
(1998) lists five stages of an unstructured observation that were originally suggested by
Silverman (p.186):
Proposing a set of general questions
Writing field notes, which are originally broad descriptive categories but which
later are developed into focused categories
Looking as well as listening
Testing hypotheses
Making broader links.
Angrosino (2005) similarly proposes the following levels of specificity that occur during
an unstructured observation study:
Descriptive observation: All details are recorded.
Focused observation: Only those activities that are relevant are recorded.
Selective observation: Specific elements of the general activities are recorded.
With any observational study, particularly one that occurs in a naturalistic setting, the
question of gaining access is paramount. Often this involves identifying gatekeepers,
key people who can provide access. Gaining the trust of those being observed is very
important in order to gain a true picture of the setting, which is why observational studies
can take a long time to conduct. Another important aspect of observation is identifying a
key informant. This is a person who is particularly knowledgeable, insightful, or
articulate and acts as a primary resource to access others.
Collection of data can be done in many ways. In structured observation, a checklist will
often suffice. In unstructured observation there needs to be some way to record the
30

natural setting. In some cases audio- or even videotaping is possible. Taping is more
likely to be possible in closed settings, such as observation of some kind of meeting or a
class. In more open settings, typical of ethnography, even taping may not be possible. In
this case, taking notes is necessary. These notes need to be as complete as possible, and
may include observations of interactions between people, activities that occur in the
setting, and other nonverbal information, as well as records of conversations. Field notes
taken on site cannot capture as much detail as is being seen. They need to be expanded
upon as soon as possible off-site since it will be easier to add details when memory is
fresh. Most researchers recommend taking several kinds of notes and dividing the
notebook page to accommodate these. These include descriptive notes (what the
researcher is seeing), researcher notes (how the researcher is feeling about what is being
seen and what ideas are developing about future research actions), and theoretical notes
(emerging thoughts on what patterns are being seen.)
The issue of informed consent and Institutional Review Board (IRB) approval is
particularly difficult for some forms of observational research. Angrosino (2005) points
out that most IRBs require clear descriptions of who will be studied and how the study
will be conducted. They tend to want to have complete protocols spelled out ahead of
time so that potential harm to participants can be identified. However, observation,
particularly ethnography, is often characterized by emerging research design and
participation by the researcher in the activities of the people being studied. Thus,
Angrosino states, ethnographers are caught in an ethical dilemma. On the one hand the
IRB requirements would have one state that the study is not intrusive, and yet
ethnography by its very nature is intrusive (although not harmful). There is no clear
resolution to this dilemma, and its solution is very local, as IRBs at different institutions
have different interpretations of the federal regulations. Researchers should be aware of
this issue and include plenty of time before the start of data collection to work with their
IRB to obtain approval for the project. Another related issue debated within the field is
the extent to which researchers make participants aware that they are being observed for a
study. This is an issue with all data collection methods, but is especially relevant to
observation. It is discussed in the section on ethics in Chapter 5.
Recent Perspectives on Observation
The issues associated with observation, as discussed by Angrosino (2005), are similar to
those for interviews. Traditionally it was assumed that the observer was neutral and could
report objectively on the setting to obtain some measure of truth in the description of
the setting. Even when it was recognized that an absolute truth was unattainable, the
procedures of observation were designed to ensure objectivity and reduce bias. These
procedures include member-checking and triangulation across multiple data sources,
which are discussed in Chapter 5.
It is now becoming recognized, however, that pure objectivity cannot be achieved. Biases
come from both the observer and those being observed. Angrosino (2005) asks the
question, Because it is likely to be the ethnographers who write up (or at least collate or
edit) the results of field studies, do they not continue to claim the implicit status of

31

arbiters/mediators of social/cultural knowledge? Those being observed also bring

differing perspectives, based on gender, ethnicity, class, or positions of power. Thus,
observation, particularly in the area of ethnography, is coming to be seen as a
collaboration between the observer and the observed. The result is a dialogue between the
two that results in a shared understanding of the setting. Ultimately, Angrosino suggests
that, It might be useful to shift from a concentration on observation as a method per se
to a perspective that emphasizes observation as a context for interaction among those
involved in the research collaboration (p. 732).

Documents and Materials

In many ways use of documents and other materials is the forgotten data source for
social sciences research. Yet many scholars find that the paper trails, as well as the
artifacts produced by people, are highly informative to their studies. While documents
consist of a variety of print (hard copy or digital) sources, other materials that may be
examined include photographs, visual art, music, instruments, clothing, or other products
that can help to tell the story.
Documents
McCulloch (2004) traces the history of document analysis, noting that the use of
documents reached its height in the mid-20th century, but has been in decline in favor of
interviews and observations since then. Nevertheless, there are instances of documents
and materials being used in engineering education research: for example, Wankat's
(2005) use of Web-based information on design contest winners and Jarosz and BuschVishniac's (2006) examination of syllabi to identify the current body of knowledge in
mechanical engineering.
Documents can be used in many ways. Punch (1998) provides four ways in which
documents can be examined:
Social production: What are the social, historical, or cultural conditions that led to
the production of the document?
Social organization: What information is intended to be conveyed by the
document? How was it intended to be used?
Meaning: What is the direct, surface meaning being conveyed by the document?
What are the deeper meanings on which the surface meanings rely?
Alternate epistemological perspectives: Use of critical or postmodern perspectives
to examine documents in other ways.
Thus, documents can be used in many different ways. They can provide historical
information that tells about how certain issues were perceived or dealt with in the past.
As in the examples cited above, they can also provide a means for understanding current
issues by collating information across several institutions. An alternate approach is
through semiotics and textual analysis, in which the focus is not only the content of the
document, but also the language used and the organization of the document (Punch,
1998). These approaches can be used to examine the underlying structures from which

32

meaning is derived, and the social context within which the documents were produced.
Punch lists the following questions that need to be considered in examining a document:
How are documents written?
How are they read?
Who writes them?
Who reads them?
For what purposes?
On what occasions?
With what outcomes?
What is recorded?
What is omitted?
What does the writer seem to take for granted about the readers?
What do readers need to know to make sense of them?
Types of Documents
A common way of categorizing documents is as primary or secondary sources. Primary
sources are generally considered to be created by individuals at the time of certain events,
while secondary sources are analyses, interpretations, or summaries of the primary
sources. However, McCulloch (2004) points out some problems with this classification.
In some cases, documents that are considered to be secondary can also be considered a
primary source in the sense that it serves as a reflection of attitudes to issues in a
particular context or period (p. 32). McCulloch also discusses autobiographies as
documents that are primary in the sense of providing an eyewitness account of events, but
also secondary in that they can include reflections and analysis of the persons own life.
Thus, the distinction between primary and secondary sources can at times be arbitrary,
and may depend on the ways in which the document is used.
McCulloch instead builds his text around three general types of documents: records and
archives; printed media and literature; and diaries, letters and autobiographies. The main
points made by McCulloch surrounding each of these document types include:
Records and Archives: Archives are repositories of records that are generally
available for public access. Archives may contain official government documents,
institutional records, or even personal documents that have been collected for
general access. McCulloch discusses some of the issues associated with accessing
archival materials. These include the use of catalogs to search for relevant
documents, the time it takes to receive documents after ordering them, and the
extent to which documents may be copied. For example, some documents may be
immediately available, while others may take days to receive. In some cases
copies can be made, while in other cases manual transcription is the only method
for recording information from the document. Documents that appeared to contain
relevant information may be limited in what they provide. Conversely, one should
consider sources that may not initially seem relevant, because they may provide
unexpected information. McCulloch provides the example of finding relevant
information on 19th century education personnel of England in Treasury
33

Department records, rather than Education Department records. Overall, one must
plan carefully for access to archives, but also be flexible in their use.
Printed Media and Literature: This type of material includes publicly available
reports, mass media, and literature. Much of this material is often created for
advocacy purposes, and so care must be taken in interpreting their contents. One
useful way of using these materials is to examine them in light of the broad social
issues that they reflect. One can use them to understand the specific social context
of a time period (past or present), or to trace broad social changes that occur over
time. However, McCulloch points out that such documents, particularly
government reports, should not be assumed to accurately portray how policies are
enacted. Rather, The influence of reports is often far less than their authors
imagine, and they may be read and interpreted in ways that are not anticipated
(p. 83). Another way in which these documents can be understood is by
examining the discourses that the document represents; that is, the particular
assumptions and arguments that are used to define a particular worldview.
Comparison across many documents can often reveal these assumptions and make
clear what is being omitted.
Diaries, Letters, and Autobiographies: All of these types of documents share the
common characteristic that they provide a personal view of events and issues.
However, there are important distinctions to be made among them. Diaries (and in
more recent times blogs) often (although not always) provide an immediate
description, with the emotions and reactions unaffected by time or reflection.
Here, as McCulloch, points out, one must careful to consider the motivation for
writing the diary or blog. In some cases it is a personal creation, not written with
the explicit intention of being read by others. In other cases, however, they are
written with the intent of being published or to espouse a certain view. This may
be more descriptive of blogs than diaries, since by their nature blogs are
immediately available for public reading. Letters are different, in that they
represent a two-way exchange between the writer and the reader. Again, there are
different categories of letters; some are official and seek to persuade or to
provide a view on a public matter. Others are more personal. However, the
distinction between these two types can be blurred. McCulloch provides several
examples of personal letters that also reveal information about development of
public policy. Autobiographies provide a personal view, but without the
immediacy of diaries of letters. Events and their interpretation are modified
through difficulties in memory, a desire to espouse a certain point of view, or an
attempt to show the events in ones life as contributing to an overall theme. Even
though autobiographies are the telling of one persons life, they can offer
interesting insights into broader social and public dimensions (p. 121).
Throughout his book, McCulloch offers many examples of how these different
types of documents can shed light on issues impacting education. Minutes of
government agency meetings, diaries and letters of political figures, and policy
documents all provide insight into the development of educational policies.

34

Minutes from local workshops are used to show how these policies are actually
enacted. Newspaper articles and even works of fiction show the broader social
context of these policies and the general view of society. This thread of education
policy provides just one example of how documents can be used. McCulloch
argues that documents provide an opportunity for connections to be made,
especially to enhance the appreciation for the historical dimension of education
and society (p. 128).
Other Materials and Physical Traces
Although not as commonly used, other materials can also inform research studies.
Participant productstheir songs, their art, their clothinghave been very important
traditional sources for ethnographic studies. Creative educational researchers have used
time lapse photography to document ways in which a room changes as students do group
work; they have examined the sculptures or drawings that participants produce to convey
their sense of an experience or idea. In engineering education research, student products
lend themselves to analysis. For example, the devices produced in the familiar egg drop
competitions might be examined to document the basic assumptions or level of creativity
of members of the group that produced them. A machine or device created in a laboratory
can be analyzed according to the criteria that it was supposed to evidence in order to
study students understanding of laboratory instructions and goals.
Merriam (1998) describes a body of literature on the use of physical traces as qualitative
data. Some examples include: comparing the answers obtained from a group of people on
how much beer was consumed at a party to the data collected by counting the beer cans
in the trash; looking at how people seat themselves in classes as an indication of racial
integration; and examining library records to see what the rate of withdrawals indicates
about the specific usage patterns of patrons. She includes cautions about the assumptions
behind these methods and advice for using these more creative data gathering strategies.
Ultimately, even though both materials and physical trace sources are not text sources,
the researcher in qualitative studies converts them to text descriptions in the study. They
are then part of the written data sources and are analyzed with other data.

Summary
Consideration of data collection methods ranges across multiple issues. From the research
design standpoint, it is important to select types of data and data collection methods that
answer the research questions and are consistent with the epistemological perspective of
the study. For example, we have seen that interviews are most appropriate for certain
types of studies, such as phenomenology, and that the specific interview questions must
also be consistent with the perspective. There are also a number of what might be called
mechanical issues associated with data collection. These include items such as whom
or what is being studied and what needs to be done to collect the data (e.g. how many
interviews or observations, what is required to access documents). Finally, there are
issues of interpretation: What biases are present in the data? How will these biases be

35

minimized or be articulated in the interpretation? Too often at least some of these issues
are not considered until data collection is underway, or even worse not until after data
collection is completed. Failing to address the issues compromises the integrity and
trustworthiness of the study. The data you collect is at the heart of your study. Without
good data you have no chance of answering your research questions, and thus it is vital
that all aspects of data collection be considered during the initial research design.

References Cited in This Section

Angrosino, M. V. (2005). Recontextualizing observation: Ethnography, pedagogy, and
the prospects for a progressive political agenda. In N. K. Denzin & Y. S. Lincoln
(Eds.), The Sage handbook of qualitative research (3rd ed.) (pp. 729-745).
Thousand Oaks, CA: Sage.
Crotty, M. (2003). The foundations of social research. Thousand Oaks, CA: Sage.
Fontana, A., & Frey, J. H. (2005). The interview: From neutral stance to political
involvement. In N. K. Denzin & Y. S. Lincoln (Eds.), The Sage handbook of
qualitative research (3rd ed.) (pp. 695-727). Thousand Oaks, CA: Sage.
Jarosz, J. P., & Busch-Vishniac, I. J. (2006). A topical analysis of mechanical
engineering curricula. Journal of Engineering Education, 95(3), 241-248.
Koro-Ljungberg, M., & Douglas, E. P. (in press). State of qualitative research in
engineering education: Meta-analysis of JEE articles, 2005-2006. Journal of
Engineering Education, 97.
Kvale, S. (1996). InterViews: An introduction to qualitative research interviewing.
Thousand Oaks, CA: Sage.
McCulloch, G. (2004). Documentary research in education, history, and the social
sciences. New York: Routledge/Falmer.
Merriam, S. B. (1998). Qualitative research and case study approaches in research (2nd
ed.). San Francisco: Jossey-Bass.
Patton, M. Q. (2002). Qualitative research and evaluation methods (3rd ed.). Thousand
Oaks, CA: Sage.
Punch, K. R. (1998). Introduction to social research: Quantitative and qualitative
approaches. Thousand Oaks, CA: Sage.
Wankat, P. C. (2005). Undergraduate student competitions. Journal of Engineering
Education, 94(3), 343-347.

36

5. Maintaining and Judging Quality in

Qualitative Research

Several scholars have called attention to the charged nature of the word rigor by
associating it with rigor mortis. They argue that the pursuit of methodological
orthodoxy (controlling all variables, eliminating all personal judgment, and the like) can
result in a lifeless portrait, useless for any practical purpose. They frequently point out
that the notion of rigor is often employed as a code word for particular methodological
preferences and thus has become more political than substantive in nature.
Nevertheless, researchers and users of research universally ask questions about quality, as
well they should. They want assurances that the research can be trusted, is useful, and
was done ethically. These questions range from the traditional concern with research
validity, reliability, generalizability, and objectivity to more overarching questions about
the impact of the research on the people involved and the worth of the research to society.
This chapter will deal with proper use of method, questions of research value, and ethics.

Proper Use of Method

Given previous chapters that describe the various research approaches and purposes that
qualitative researchers employ, it is difficult to fully address questions of methodological
soundness universally. However, there are some overarching concerns that researchers
everywhere ask about the quality of research.

Was it designed well?

Were appropriate methods used consistently?
In what ways did the researcher influence the results?
Can the findings be applied to other settings?

These questions have been addressed by scholars of qualitative research in several

fashions over the years. At first, attempts were made to identify the equivalents of the
traditional terms, validity, reliability, objectivity, and generalizability. Lincoln
and Guba (1985) are frequently cited for describing equivalent goals and the ways in
which qualitative researchers might achieve them. Their arguments are summarized
below.
The concept of internal validity as the concern with capturing what is really present in a
research study undergoes some modification in a paradigm that is based on multiple
realities. Lincoln and Guba (1985) refer instead to credibility. They ask whether the

37

researchers findings are plausible; that is, would reasonable people believe that they are
accurate descriptions? In order to achieve credible findings, they advocate the following:
Use prolonged engagement and persistent observation. Stay in the field long
enough and look at as many cases as needed to thoroughly know the phenomenon
of interest.
Triangulate (compare results of) different methods, sources, and theories. Capture
multiple points of view, use different approaches, and entertain different
theoretical perspectives to look for patterns that persist across all of these.
Triangulate researcher perspective using peer debriefing. Confer with colleagues
about methods and data interpretation to ascertain whether approaches are
reasonable and appropriate.
Triangulate with participants using member checks. Interact with research
participants or those who are contributing data to ask whether your data recording
and interpretations are accurate from their points of view.
Collect referential adequacy materials. Seek out descriptions, quotations,
documents, or products that provide the support for your interpretations.
Use structural corroboration. Make sure that your arguments hang together and
make sense, that they are logical and describe the data well.
Reliability is the concern with consistency of the methods and findings. Its parallel in
Lincoln and Gubas (1985) scheme is dependability. They advocate that researchers
use overlap methods, by which they mean triangulation of method as described above,
and stepwise replication, the systematic checking of research processed during the
study. They also suggest that researchers leave an audit trail, describing their
approaches carefully and detailing what inferences they made in drawing their
conclusions. An external auditor should be able to look at the evidence and make a
judgment about the dependability of the research.
Objectivity is another of the traditional criteria that requires reconceptualization in a
multiple-truths paradigm. The concern is with researcher bias and there are several
approaches to this issue. The most common is to have colleagues of the researcher, either
through the triangulation or audit approaches described above, examine the work for bias.
A more fundamental approach is to assert that all research is conducted from the position
of the researcher and so is inherently subjective. More conservative scholars suggest
that by describing their positionality thoroughly, researchers can control their bias
through the use of devices such as reflexive journals, surfacing to conscious awareness
their emotional reactions and taken for granted assumptions. They also recommend that
the researcher alert the user of the research to the point of view through which the
research was conducted by describing this fully in the research report. Less conservative
scholars argue that ones positionality is a source of strength and is an essential part of
the research. They agree that this perspective should be described as openly as possible
for the user of the research but not from the point of view of trying to control ones
subjectivity, but to use ones life experience and judgments as an important research tool
and to reveal this perspective as well. In critical approaches, an identified lens, such as
Critical Race Theory, Feminist Theory, or Marxism, is an essential part of the method,
which does not pursue an unbiased view, but rather an identified one. The avowed aim of

38

this research is to produce change rather than to capture reality independent of a knower,
which is not believed possible.
Generalizability or external validity is the concern with whether the findings of the study
describe all cases for which it is designed to apply. Lincoln and Guba (1985) substitute
the term transferability, acknowledging the hope of all researchers that their findings
will inform other cases in addition to the one they described. The goal of a particular
study is not to provide generalized findings that apply in all contexts, but to provide a
description that applies within the context being studied. Again, a paradigmatic
difference enters within the context of multiple realities that contradicts the notion of
absolute generalizability. The shift in qualitative research is to locate ultimate
responsibility for applicability with the user rather than the researcher. That is, people
seeking to know how applicable given research findings are to their cases of interest must
estimate the match between the context of the research and their own cases in order to
make this determination. The researcher can help by using what scholars call, thick, rich
description, portraying in extensive detail the particulars of the context of the research,
so that the user of the research can make the transfer.
Creswell and Miller (2000) summarize the steps that a researcher can take to assure
methodological soundness. They recommend that researchers employ at least two of
these approaches in a given study. Many of the steps echo the recommendations of
Lincoln and Guba (1985) above:

Triangulation: As described above, use multiple methods, sources of data,

theoretical explanations or investigators to collect or analyze data. Look for
convergence across these.
Divergent cases: After generating preliminary themes or explanations to describe
the data, seek out cases that do not fit and either modify or reject the original
explanations.
Reflexivity: Describe your bias to potential users of the research. Try to either
suspend your own perspective or articulate your assumptions and position in a
commentary or prologue to the research.
Member checks: Ask participants in the study to confirm the accuracy of the data
you have recorded or to respond to your interpretations.
Prolonged engagement: Collect enough data to provide the basis for sound
interpretations. Stay in the field for ample time, interview enough participants,
and analyze sufficient documents.
Audit trail: Provide evidence on how you conducted the study and formed your
interpretations. This might be in the appendix (such as the interview questionnaire
or the coding framework) or incorporated into the presentation of findings.
Peer debriefing: Ask a colleague to review your decisions and interpretations for
their reasonableness.
Thick description: Use details to describe context and people in the study so that
the reader can have as much information as possible to understand the findings.
Illustrate the narrative of a qualitative study with images, quotations, and colorful
descriptions.

39

Collaboration: Work with research participants as co-researchers through

generating research questions or research design together, by teaming to collect
data, or by jointly interpreting data.

Issues of Responsibility and Ethics

In addition to proper use of method, scholars of qualitative approaches identify other
hallmarks of study quality that are less frequently discussed in the traditional research
literature. For example, Lincoln (1995) lists the following: locating the research in and
serving the purposes of the community in which it was carried out; giving voice to the
participants through direct quotation rather than filtering through the researcher; and
reciprocating through sharing trust, privileges, and ownership of the study with the
participants. Other scholars discuss usefulness as an important criterion. If a study is
trivial or has little potential to inform or change human behavior or enable deeper
understanding of the human condition, it fails the quality test, even if it uses methods
soundly.
Major ethical issues that are also associated with the quality of a qualitative study include
the following:
Informed Consent: The participants should know what the research is about and how
information will be shared and used. Their consent should be renegotiated if the
design or usage arrangements change.
Confidentiality: Participants' identities are only revealed to agreed-upon extent
(usually not at all) and the findings and data are communicated in such a way that
these identities are protected. Pseudonyms or codes are used in records and records
are kept secure.
Avoiding Harmful Consequences: Research manipulations, interventions, or
suggestions should not threaten the well being of participants. The researcher should
not use his or her authority as a platform for influencing participants thinking.
Genuine Reasons for Conducting Research: The decision to undertake research
should stem from sincere interest in the topic or commitment to change. The
researcher should avoid pursuing trendy topics or fundable projects only out of
opportunism rather than sincere belief in the importance of the study. Researchers,
especially in evaluation contexts, should avoid being co-opted by those in control.
Honesty: The researcher should avoid forcing conclusions, fudging data, or other
clearly dishonest approaches.
Reciprocity: Responsibility for respecting the contributions of participants at
minimum involves reporting back findings. In addition, some researchers feel that
they should share the royalties or other benefits accruing from the study with the

40

participants or seek to promote change in the setting as a result of the new

understandings.

Paradigmatic Issues
As the above discussion indicates, paradigm issues in the conducting of research figure
quite significantly in discussions of research quality. Prevailing quantitative standards
can seem to be at odds with the purposes and assumptions of qualitative research, which
can put the researcher in a somewhat defensive position. Patton (2002) articulates some
stances that can help to bridge the gap in expectations by focusing on the different goals
of research traditions. Three of his main points include:

Stress fairness rather than researcher distance. Instead of claiming that the
researcher is absent, dispassionate, or totally without perspective, assure readers
that the researcher used all his or her intellectual and methodological tools to
describe the focus of the research in a way that was fair and reasonable. This can
be seen as a strength, compared with traditional studies in which the reader knows
nothing of the researcher stance and how it affected the use of methods. As Patton
says, Distance does not guarantee objectivity; it only guarantees distance (p.
575).
Stress promeaningfulness rather than antinumbers. Although numeric data
have a credibility in our society that is often way beyond their intended strength
or capacity, emphasize the strength of text rather than attacking numbers. Focus
on difference, not deficiency, concentrating on showing how the approach was
able to illuminate the really important questions of the study and how it produced
compelling results.
Stress usefulness rather than truth tests. Instead of engaging in debates that are
predicated on the assumption of absolute truth, focus on the usefulness of having
actual portraits and insights into the perspectives of research participants for
understanding present realities and determining future actions. Emphasize the
pragmatic nature of qualitative findings and their utility and compelling nature for
those who use them. Although it is often said that decision makers want hard
numbers, a powerful story frequently makes a more lasting impression and
engenders quicker response.

Summary
Conducting sound qualitative research should be the primary goal of all who use these
approaches. Although the usual definitions of terms referring to the quality of research
are often embedded in philosophical assumptions that are not shared by qualitative
researchers, scholars have been able to articulate standards of good research that apply to
qualitative studies. Strategies for achieving high quality research have been identified.
Researchers should employ these in their work and users of research should seek
evidence that they have been employed in determining the quality of the studies that they
encounter.

41

References Cited in This Section

Creswell, J. W., & and Miller, D. L. (2000). Determining validity in qualitative research.
Theory into Practice, 39(3), 124-130.
Lincoln, Y.S. (1995). Emerging criteria for quality in qualitative and interpretive
research. Qualitative Inquiry, 1, 275-289.
Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Newbury Park, CA: Sage.
Patton, M. Q. (2002). Qualitative research & evaluation methods (3rd ed.). Thousand
Oaks, CA: Sage.

42

6. Data Analysis
Novices to qualitative research often experience delight during data collection. Although
they may encounter glitches along the way, the work of observing, interviewing, or
examining documents or products is engrossing and often seems easier than expected.
When it comes to analyzing the data, however, qualitative research often throws novices
into a quandary. How can all the pages of text that have accumulated be sorted in an
organized fashion? What strategies should be used to interpret and communicate the
patterns that are present?
Although there are many ways to analyze qualitative data, and in the end, the strength of
the approach relies on what the researcher brings to the table by way of personal insight
and experience, the steps that follow outline a general approach. After that, some
strategies for particular types of qualitative studies are included.

Begin at the Beginning

The activities of qualitative research are more recursive than linear. Since the design is
flexible rather than fixed, the researcher begins analysis during data collection, rather
than collecting data first, then turning to analysis. This work, called interim analysis,
starts from the early days of data collection and escalates gradually so that as the study
proceeds more time is devoted to data analysis and less to collection. Reflective notes
that record the researcher's feelings and thoughts about the data help by stimulating ideas
during analysis and by documenting possible bias. Analytic notes are used for
hypotheses, interpretations, and growing realizations about the data. Examples of interim
analysis activities include writing analytic memos about methodological choices, patterns
that are noted, reactions to what is being collected, or the like. Entries might be of the
following sort:

Should I also talk to the lab instructor about this issue? She might offer a different
perspective than what Im getting from the students.
Does the level of confusion in this course appear to be related to which approach
was used in their introductory sequence?
It looks like the ability to picture oneself as an engineer might be related to
personal experience with an engineer during the precollege years.
It appears that students who are more convergent thinkers are having more
problems with this than divergent thinkers.

These notes to self are the beginnings of analytic thinking about the study and are critical
in directing the researcher toward additional data collection activities, literature, or
questions to be asked of the data.

43

It is a good habit to review one's data daily during collection to make sure raw notes are
edited into a form that will be usable even after time has dimmed the memory of the
actual experiences of the day. Expanded notes that add the detail that there was not time
enough to note while collecting data are written frequently by researchers in the pauses
between data collection activities during a study.
A second activity that needs to take place right from the start of data collection is data
storage and organization. Given the large volume of text that normally accrues from a
qualitative study, even the most casual researcher has to adopt structured habits for
labeling data in standard fashion (who, when, where, pagination, file name, and the like)
and keeping it safe (according to Human Subjects confidentiality agreements) and with
like pieces (such as in binders, electronic file folders or drives, or in labeled paper folders
in a locked file cabinet).

Reviewing and Selecting Data

As data accumulate, the more the researcher reviews them, the easier subsequent work
becomes. Some of the data collected in a qualitative study, especially the data collected in
the early period of a study before the focus was narrowed, will be peripheral to the main
research questions. Interview transcripts will contain whole pages that document times
when the interviewee strayed from the subject or digressed into an amusing but not
relevant memory. Field notes will often contain laborious descriptions of the actions of a
participant who later dropped out of the study. These data segments are candidates for
what some researchers call data reduction.
Deciding what data are most pertinent to the study, however, is not always
straightforward and requires the judgment of the researcher. Like all other analysis
activities, selecting useful data from the total data collected is an activity that depends on
the care and sensitivity that the researcher brings to it. Researchers vary in their selection
decisions, some finding that they can reduce the amount of data they analyze at an early
stage without damaging its integrity, and others feeling that they must work with a larger
amount of data through the whole process. While data not considered relevant at the start
of analysis can be set aside as other data become the main focus, they should not be
destroyed in the event that their importance only becomes clear as the analysis proceeds.

Sorting the Data into Categories

As data are selected for further analysis, the researcher develops a sense for the ways in
which they can be fitted together. Often, a preliminary sort can be made on the topics that
were explored through the observations or interviews. These may correspond to interview
questions when structured interview formats are used. For example, all comments in the
interviews on first-year students who abandoned engineering as a field of major study
might be placed in such preliminary categories as perception of lack of fit, academic
issues, time management, financial, or the like. Selecting a comment is called
chunking, and deciding how much or little of a given text should be selected for
classification is a decision made on the basis of the type of study. For standard studies (as

44

opposed to microlinguistic studies when each word is scrutinized), one to five sentences
are usually taken to be a chunk; that is, an appropriate unit to code.
The category names are attached to the data passages in some way, either as marginal
annotations, labels on cards on which the passages have been pasted, or category fields
on a database record. Often, the category names are abbreviated for this coding process
(fit, academic, time, money), but they should not be so shortened that the
researcher has a hard time remembering what the label stands for. This sorting may be
fairly straightforward. Frequently, however, the decisions become more complexthe
researcher realizes that some categories seem to be a blend of others, such as academic
issues may include prior preparation, instructor conflicts, or level of challenge, in the
case above. Some comments or notes or document passages seem to belong to more than
one category. The distinctions between the categories become blurred. New categories
seem necessary after much of the data have already been sorted. Frustration sets in.
Some of this is inevitable, but can be alleviated by thinking of sorting as an iterative
activity that continually requires refining categories until they accommodate the data.
It helps to first sort through a part of the data in an exploratory fashion, simply writing
down a list of category names that seem useful and continually adding to, subtracting
from, or modifying this list. Next to each of these names a definition of the boundaries of
the category can be written that also can be modified as the researcher proceeds. Such
notations help with decisions as to whether a piece of data fits into that category. This
preliminary list of categories and their definitions will almost always change as analysis
proceeds, but it can form the core of the category system that will eventually be used.
Researchers call this a coding taxonomy or a codebook. Depending on the complexity of
the study, the list of categories might resemble a tree structure. For example in the study
above, there are several kinds of academic issues, which are in turn an attrition issue.
When more than one researcher is working on a set of data, more formal use of a
preliminary list of codes will be necessary. Researchers then code the data and compare
and reconcile coding for differences.

Managing the Data

There are several different ways forf indicating selection of data and subsequent labeling
of the data. Some researchers use highlighting pens or stickers affixed to note pages or
transcripts; some pencil in bracket marks around the selected passages; some cut selected
portions of the data from copies of the notes, transcripts, and documents and paste these
on cards; some enter the passages into a computer database management record. Standard
database programs such as Access or Filemaker Pro and spreadsheets such as Excel can
be helpful. Specialized qualitative research analysis software packages such as NVivo or
Atlas are also available, relying on tags applied to text strings to store data. The latter are
the preferred method for a study with large amounts of data. With any method, care must
be taken to be able to trace text excerpts back to their original context. Cards and
database records should contain transcript numbers, page or line numbers, or other
identifiers that will enable the researcher to trace the passage to its original context easily
as analysis proceeds. Specialized software does this automatically.

45

Using database software for data management involves the following:

Creating fields for, at minimum, the ID number and location of the segment, the
category of the segment, the theme(s) represented by the segment, and the text of
the segment
For each coded segment, pasting the selected text from the original document to
the text field
Filling in the ID number and category name for each segment (the themes will be
entered later)

For example, two records might look like this (the theme field will be discussed later):
ID
7

Location
Transcripts,
p. 118

Category
Time

Theme
Selfpacing

Text
In high school, my parents made me
study at night, but here, I had to decide
when and how much to study and I got
caught up in other things and never got
around to it.

Transcripts,
p. 87

Fit

Critical
mass

I looked around and there wasnt

anybody like me in the class or really in
the whole engineering school.

Essentially, an electronic grid of numbered segments (records) is being created for

storage and sorting. Spreadsheets can serve the same function, although special database
programs are more suited for more complicated text manipulations. Selecting the
memo option during creation of the text segment field in Access expands the amount of
text allowable in the field.

Displaying the Data

Even when the data have been reduced in quantity and coded carefully, it is hard to
develop a sense of the patterns and meanings they suggest until they can be visually
scanned in their categories. Some researchers cut and paste the coded data at this point on
to long sheets of paper or group together index cards on which they have pasted the
segments of interest; those using computers sort their records and visually scan them or
obtain printouts by category. In the example above, separate printouts or screen displays
of each category (time, fit, money, etc.) would be made, allowing the researcher to look
at all the text segments that have been placed into this category together. Ways of
displaying the data will vary according to the resources that the researcher has, the type
of data and study, and the amount of data on hand. Through looking at the assembled
segments, the researcher can decide to change the category label in the field, to move
some segments to other categories where they fit better, to split one category into more
than one and recode the segments accordingly, or to rename to category or readjust its

46

boundaries to better accommodate the data. Data analysis in qualitative research is

recursive.

Reviewing Within and Across Categories for Themes

Once data are grouped into formats that permit scanning, they can be reviewed for
common ideas or "themes" that they contain. In the hypothetical study about attrition, for
example, when all the comments about time management are arrayed together, the
researcher may find that in one way or another, almost all of the respondents said that
their newfound autonomy and the distractions of living with others their age led to poor
choices on time management (a label called self pacing might be used to convey this).
Themes are not always stated explicitly, but labeling the main idea or ideas in a text
segment helps the researcher organize and capture the sense of the data. The themes
generated during analysis form the basis of the findings of the study. In complex data
sets, themes are labeled and coded just like categories so that instances of the theme in
the data can be retrieved and displayed. Each segment can be labeled with more than one
category or theme so that it will be retrieved with multiple sorts.

Synthesizing the Information

Making sense of the organized data involves summarizing the information either by
frequency (Self pacing was a problem of all but two students who left engineering.) or
itemizing (The main reasons why students experienced academic issues were) or
description (Women, first-generation students, and Black and Hispanic students provided
the following examples of the ways in which they felt they did not belong) As Miles
and Huberman (1994) illustrate, making grids and charts can be helpful at this point.
Often, this activity involves testing hypotheses that are formed during the analysis against
the evidence to see if claims are warranted (Did students who lived on campus have more
problems with time management than those who lived off?) Negative cases must be
accounted for in explanations or descriptions that are formed (the off-campus student
who had no time management issues). Often, the original hypothesis must be modified or
discarded because support from the data is lacking.
At other times, qualifying or contextual data must be considered in interpreting the
meaning or strength of a given finding. Fitting the pieces together involves
"triangulation" (testing data against each other), building a logical chain of evidence, and
"structural corroboration" (making sure that the picture of the whole that is portrayed
makes sense, is supported by the pieces of evidence that constitute the finding).
Whenever possible, the researcher should check his or her interpretations with the
original respondents or colleagues for support or disconfirmation.
When data are organized by category and theme, an outline of the presentation of
findings can be derived. As each topic is discussed, the arrays of the particular category
or theme being treated can be used to supply examples, quotations, or frequencies to add
detail to the report. Charts and tables made from the arrays can also be helpful in the final
report.

47

Specialized Approaches to Analysis

While the steps above describe a basic approach to text analysis (also see Richards, 2005,
for a detailed description), different traditions within qualitative research have developed
specific ways of handling data analysis. Creswell (2007) does an excellent job of
summarizing these. Notes on a few types of differences are provided here.
Case Studies: In embedded case studies, in which there are cases within the case
(for example, students within a class), each case is analyzed separately and then
themes across the cases are analyzed.
Critical Studies: In addition to doing the usual thematic analysis, the researcher
in a critical study analyzes the data in terms of the critical perspective that is
taken. For example, in a cultural reproduction study, the researcher would
undertake a special examination of the data for instances in which the existing
power structure is being supported by the actions of those in control. Some critical
theorists believe that inherent in the analysis activity is the responsibility for the
research to produce actual change leading to informed action (praxis) in the
setting.
Phenomenological Studies: In phenomenological analysis, the researcher first
examines his or her own experience with the phenomenon of interest to separate
this out, as far as possible, from that of the participants. Next is a listing of the
significant statements made by the participants about how they experience the
subject of the study (horizontalization). These are then grouped into broader units
of meaning and summarized with examples of what was experienced (textural
descriptions) and how it occurred (structural descriptions). A brief composite
statement is then written that summarizes the essence of the experience.
Moustakas (1994) is a source for additional details.
Grounded Theory Studies: Grounded theory studies also involve a multistage
process. Line-by-line open coding (similar to the description above) first occurs,
followed by dimensionalizing, which calls for arraying the data along a
continuum, and axial coding, which engages the researcher in drawing logical
connections between a category and its subcategories. The researcher then creates
a conditional matrix, a diagram that maps the range of conditions and
consequences and then performs theoretical sampling, testing emerging
hypotheses against the cases. Finally, the researcher integrates the conclusions
into a theory statement. Charmaz (2006) offers a modern perspective on analysis
in grounded theory studies.

48

References Cited in This Section

Charmaz, K. (2006). Constructing grounded theory: A practical guide through
qualitative analysis. Thousand Oaks, CA: Sage.
Creswell, J. W. (2007). Qualitative inquiry & research design: Choosing among five
approaches (2nd ed.). Thousand Oaks, CA: Sage.
Miles, M. B., & Huberman, M. (1994). Qualitative data analysis: A sourcebook of new
methods (2nd ed.). Thousand Oaks, CA: Sage.
Moustakas, C. (1994). Phenomenological research methods. Thousand Oaks, CA: Sage.
Richards, L. (2005). Handling qualitative data: A practical guide. Thousand Oaks, CA:
Sage.

49

7. Writing Up Qualitative Research

As a Professional Engineer, I dedicate my professional knowledge and

skill to the advancement and betterment of human welfare (NSPE, 2008).
Qualitative research and the written dissemination of findings are natural extensions of
the idea that an engineers role is to serve humanity. Qualitative writing within this
context not only recognizes the humanity of the participants or phenomena studied but
deliberately and conspicuously reflects it.
The effective reporting of findings, whether quantitative or qualitative, involves the
accurate and complete reporting of meaning. Nevertheless, there exists a difference in
how this meaning is constructed and ultimately relayed depending on the research
approach. Richardson and St. Pierre (2005, p. 960) describe this distinction: Unlike
quantitative work that can carry its meaning in its tables and summaries, qualitative work
carries its meaning in its entire text. Just as a piece of literature is not equivalent to its
plot summary, qualitative research is not contained in its abstract. Qualitative research
has to be read, not scanned; its meaning is in the reading.
The goals of this chapter are to: 1) introduce several overarching, rhetorical issues related
to qualitative writing of any kind, and 2) present a general organizational structure as
well as specific (or embedded) writing devices/techniques associated with each of the six
major approaches to conducting qualitative research presented in Chapter 2.

Overarching Rhetorical Issues in Qualitative Writing

So what makes qualitative writing so different? First, there exists a major shift in
paradigm from the more traditional reporting associated with technical research to the
purposeful recognition of and reflection on the voices of both the researcher and the
participants in a given study. Intuitively, the typical engineering researcher understands
that people are complex and cannot be calibrated. Concurrently, every variable associated
with the lives and experiences of the engineering professor charged with developing a
new curriculum or 8th graders attending their first engineering summer camp cannot be
controlled. Qualitative writing recognizes the complexity of the human experience and
acknowledges, as Creswell (2007, p.179) suggests, ...the impact of the writing on the
researcher, on the participants, and on the reader.
Qualitative writing is contextual in the sense that the researchers shape the writing based
on their own interpretations, background, or experiences (Creswell, 2007). The authors
level of reflexivity, meaning their ability to recognize, accept, and write openly about
their perspectives, is a key component to reporting qualitative findings effectively.

50

Related to issues of author reflexivity are those associated with how participants in the
study are both impacted by and ultimately impact the research. Gilgun (2005) contends:
Any form of science requires that researchers do careful observations and
represent observations as well as they can. Social scientists observe, interact with,
transform, and are transformed by other human beings. Thus, we social scientists
have the task of figuring out how to represent ourselves and other human beings
in the most full and accurate way possible (p. 260)
Consequently, it is prudent for the researcher to ensure that the writing reflects the unique
voice of the participant. This critical examination of context also extends to the reader
who may form an entirely different interpretation of the findings, different from both the
author and the participants (Creswell, 2007).
A second rhetorical issue to be considered is the audience for whom the report is written.
The key idea is to purposefully reflect on who may ultimately read your work and extract
their own meaning from your findings. The goal is to present your findings in such a way
as to maximize their usefulness to readers (Rossman & Rallis, 2003). In most cases, your
audience may be other scholars who read and submit to refereed journals; however, your
audience may also consist of engineering department administrators, student affairs
professionals, or even the parents of current students who all want to understand the
value, as they each uniquely define it, of the new engineering service-learning initiative
or study abroad program you are exploring.
Last, the language used to shape qualitative writing must also be considered. How
researchers encode qualitative reports reflects both their perceptions of the findings as
well as what they perceive to be of value to the reader (Creswell, 2007). For example, if
you perceive the voice of the participants in the study to be of particular value to helping
the reader find meaning in your findings, then you may consider including various direct
quotes from the participants. Conversely, if your study involves working with
engineering faculty and you happen to be faculty member, it is likely that including your
own insights from this shared point of view will add a critical depth and richness to your
report.
As mentioned previously, it is also important to consider your audience as you
contemplate the language of your text. When writing for the academic community, for
example, it is common to use terminology, metaphors, or images inherent to that
particular field. Writing for non-academic or lay audiences may require language that
reflects a greater focus on practical application and less concern with research methods
and the like (Richardson, 1990). Even in cases where a more structured format is
required, the language of qualitative writing reflects these various paradigm shifts as
methods become procedures, results are now findings, and subjects are now participants
(Creswell, 2007).
The important point is that qualitative writing should reflect the complex and iterative
interactions that exist between the author, the participant, and the audience throughout the
study. Qualitative research and the subsequent reporting of findings require an intense,

51

critical examination of myriad data and the ability to make and communicate meaning.
Lichtman (2006, p. 178) contends that to do this effectively, the researcher must develop
a structure that fits the data you have collected, the ideas you are trying to convey, and
your personal style.
In summary, some characteristics of the writing in qualitative studies are different from
traditional studies, although as will be described in the next section, there are variations
and exceptions. The main differences in general are that qualitative studies:

May have different divisions. The usual sequence of introduction, justification,

literature review, definition of terms, description of methods, presentation of
results, and discussion may not be present or may be included in a different order.
An opening may begin: Jogi Singh, a first-year mechanical engineering student,
groaned when he heard that his Principles course would revolve around
collaborative learning groups. There is always a person who does the work (me)
and the others who wait for me to pull them along, he complained.

May be written in first person. The researcher is not trying to imply that he or she
had no active role in the study. A statement in the methods section might read: I
had always resisted group work myself, so I was curious about the reasons why. I
set about to explore this by

May be tentative about literature, theory. The focus may be more on the
description of a particular case or cases.

Usually do not define terms, operationalize variables, or list hypotheses.

Include discussion about the specific tradition of inquiry and why it was chosen.
This discussion might begin: I chose an ethnographic approach for this study
since I wanted to describe the culture of the practicum experience at this
university. Questions I wanted to ask centered on the rules of the gamehow does
one get a good placement? How must one operate in order to be thought a sharp
prospect? . . .

Articulate the rationale for purposeful sampling. This may be expressed in terms
such as the following: I chose maximum variation sampling because I wanted to
see how the virtual laboratory was experienced by students who differed in
gender, prior GPA, and comfort with technology (as measured by our earlier
survey). It seemed to me that these factors might influence their use of the lab
because prior studies {citations} have talked about these as three key factors.

Provide more information about the context of the study. For example: The
students in the sample were all first-generation males who were recruited through
a four-week summer program. This program has had a history of innovation. Its
director talks about the ways in which it has changed [quote]. Support for the
program had varied, depending on the college leadership over the years [details]
...

Spend more time treating ethics, trustworthiness, and reflexivity.

52

Include more description about analysis procedures.

Present quotations, examples, displays, or other documentation that make the

report come alive . . . and often cause it to be longer than traditional reports.

Overall and Embedded Rhetorical Structures for Six Qualitative

Approaches
The varying levels of structure and flexibility associated with reporting ones findings are
dependent upon the qualitative research design being employed. Below are specific
considerations for the six major approaches described in Chapter 2.
Narrative Studies
The process of writing up narrative research is essentially an exercise in experimentation
with flexibility and form (Clandinin & Connelly, 2000). Identifying your own narrative
form involves thinking about your own personal preferences as a reader or reviewing the
narrative works of other authors. As Chapter 2 described, narrative research is interested
in the life story of an individual or individuals. Depending on the individuals, their ability
to recall specific life events or the ascribed importance they may place on certain events
over others, the telling of these stories may transcend any specific boundaries related to
time (e.g., present versus past tense) or theme (e.g., childhood school versus home life)
all within one interview session.
Several kinds of narrative writing structures are available to the researcher to complement
this inherently dynamic process. In a more structured approach, the author is free to
disseminate the participants story chronologically. This approach is likely to work best
when you are attempting to show some sort of personal development on the part of the
participant over time. Conversely, a more flexible and widely used approach is the threedimensional inquiry space model (Clandinin & Connelly, 2000). This narrative
framework mirrors the story-telling in the sense that the author is free to report on life
events without strict regard to time (e.g., looking backward and forward) or point of view
(e.g., looking inward and outward between the story as told by the participant and the
authors interpretation of it) while situating the story within a specific context.
Creswell lists several widely used writing strategies to facilitate the development of a
narrative text (2007, pp. 184-187):

The author is free to give more space and emphasis to certain participant stories
and voices than others.

The author may use the progressive-regressive method, in which she or he begins
with a key event in the participants life and proceeds to work forward and
backward from that event.

53

Writing may emphasize certain key events or epiphanies defined as interactional

moments and experiences that mark peoples lives. It is important to note that
these epiphanies may be the result of the participants reliving the experience
while speaking with the researcher.

Narrative reports may emphasize certain themes that inform participants life
development.

Devices that help create and facilitate a smoother, more creative flow to the text
consist of transitions in the form of words or phrases or even time shifts.
Foreshadowing is often used to provide hints of events or themes to come later.
Finally, metaphors may be employed as a means of richly describing people,
places, time, and scenes, or to promote greater understanding for the reader.

Case Studies
Merriam (1998, p. 238) contends that . . . the major point about case studies to keep in
mind is that they are richly descriptive in order to afford the reader the vicarious
experience of having been there. Case study narratives are inherently and purposefully
substantive as the author attempts to paint as complete a picture as possible of what has
taken place. Although there is no standard format, researchers have provided workable
models of reporting. Lincoln and Gubas substantive case report format, as summarized
in Creswell (2007, p. 196), prescribes a thorough explanation of the problem or issue; a
detailed description of the context and setting; a description of the interactions or
processes observed within the setting; identification of saliences (elements studied in
depth) at the site; and elaboration of outcomes of the study.
Concurrently, a case study narrative may also be informed by both your research design
and unit of analysis. Yins (2003) 2 x 2 matrix describes instances where single or
multiple-case designs may be used when the unit of analysis is either holistic (single unit)
or embedded (multiple units). As the researcher, it your responsibility to discern which
approach to use depending on your research questions and purpose of the study. These
decisions ultimately affect the vantage point of the text and the subsequent level of
engagement of the reader in the case. In most cases, vignettes are used to introduce the
case to the reader as well as provide final points of emphasis to take away from the case.
Creswell lists additional devices to properly develop a case-study narrative (2007, pp.
184, 196-197):

Authors may use a funneling approach to describe the context and setting of the
case. The reader is taken from a broader picture (e.g., a mechanics of materials
course lecture) to a narrower one (e.g., a specific engineering student attending
that lecture).

54

Researchers must also decide how much description to provide in the text in
comparison to the amount of analysis and interpretation. Merriam (1998) suggests
a 60/40 or 70/30 proportion of description versus analysis.

Ethnographies
Ironically, many ethnographies are written in the tradition of the third-person, relatively
detached mode of report writing, or realistic tales (Van Maanen, as described in
Creswell, 2007). However, more researchers are exploring alternative ways of reporting
on a studied culture. Van Maanen, summarized in Creswell (2007, p. 192), describes the
confessional tale, in which the author focuses more on the fieldwork experiences than on
the culture. Alternatively, the impressionistic tale is seen as a personalized account of
fieldwork in dramatic form and exhibits elements of both realistic and confessional
writing while maintaining a first-person point of view.
Wolcott (1994) also offers a framework for quality ethnographic writing. He contends
that a description of the culture can be provided using any number of approaches as a
starting point, such as a chronological ordering of events, an analytical framework, or a
story told through several perspectives. Data analysis follows this detailed description,
including, but not limited to, the reporting of fieldwork procedures, identifying thematic
patterns in the data, and/or critiquing the research process itself. Lastly, a thorough
process of interpretation takes place during which the researcher is free to extend the
analysis, make inferences from the data based on existing literature, or interpret findings
within the context of his or her own experiences as an investigator.
It is apparent that ethnographic writing allows one to adhere to some sort of systematic
approach to a greater extent than the basic interpretative, narrative, or case-study
approaches. Emerson, Fretz, and Shaw, as summarized in Creswell (2007, p.193), suggest
that the development of an ethnographic study can best be described as a . . . thematic
narrative . . . constructed out of a series of thematically organized units of fieldnote
excerpts and analytic commentary.
The following suggested techniques for ethnographic writing were drawn from
Creswells (2007, pp. 184, 194-195) summary of embedded rhetorical structures related
to the ethnographic approach to qualitative inquiry:

Ethnographers often use tropes, or figures of speech, in their writings. Examples

are metaphors; synecdoches, in which authors present examples, vignettes, and/or
illustrations; and storytelling.

Denzin (summarized by Creswell) talks about the importance of thick description

in presenting the detail, context, emotions, and complex relationships of the
culture being studied. The goal is to facilitate a strong empathetic response in the
reader by painting as complete a picture as possible.

55

Reporting findings in an ethnographic study usually includes myriad examples of

dialogue normally written in the dialect and natural language of the culture.
Concurrently, members of the particular group are often shown talking and
relating to each other using characterization.

Ultimately, a good ethnographer is a good storyteller. Consequently, evocative

literary techniques are used to enhance the readers experience. Several examples
are the flashback, interior monologue, or dialogue, to name a few. Researchers
use many literary techniques.

Grounded Theory Studies

As Chapter 2 stated, grounded theory as a qualitative research approach originated from a
desire to incorporate a more analytical, systematic process of theory development when
conducting qualitative research. Over time, the original framework offered by Glazer and
Strauss (Creswell, 2007) has been reinterpreted and modified; however, the primary
focus of grounded theory research as a means of generating a theory to adequately
explain a process or phenomena has not changed. Charmaz (2006, p. 151) argues the
potential strength of grounded theory lies in its analytic power to theorize how meanings,
actions, and social structures are constructed. Charmaz further states thatgrounded
theory preserves and presents the form and content of the analytic work (p. 151),
suggesting that the presentation of the theory should possess structural similarities with
the process of its development.
The writing suggestions provided by Corbin and Strauss (2007) coincide with this
interpretation. They suggest the development of a clear analytic story originating in the
selective coding phase of the study. Writing is to be kept on a conceptual level, void of
highly detailed descriptions of the phenomenon itself and focused more on analytic
theory at an abstract level. Purposeful attention is paid to specifying relationships
between categories uncovered during the axial coding phase of analysis, when the
foundation of the theory occurs. Finally, the writer denotes possible variations and
different conditions under which the theory holds. Throughout this process, the narrative
may reflect changes in methodology in response to new data.
Creswell (2007, pp.184, 191-192) presents several widely used embedded writing
techniques that exist when reporting on findings from a grounded theory study:

The extent of data analysis can play a significant role in influencing what is
ultimately addressed, and to what degree, in the narrative. Some researchers may
focus on the process of linking categories around a core category in axial coding
while others may write significantly on the subsequent generation of their formal
theory.

Another embedded structure is the examination of forms used in stating the

propositions or theoretical relationships derived from the study. It is possible to

56

relay these propositions in a narrative (or discursive) form or by simply providing

a bulleted series of statements or hypotheses.

Segments of data may be represented in the form of vignettes or quotes as

explanatory tools and should demonstrate how well the theory is grounded in the
data.

Finally, as is customary in the narratives of many grounded theory studies, a logic

or integrative diagram is included to visually present the theory. The structure of
this visual model is identified during the axial coding phase and through the use
of arrows, descriptive, and/or qualifying statements, illustrates flows of causality,
context, intervening conditions, and consequences related to the theory. This
representation is usually presented at the end of the report and may be coupled
with a narrative version.

Phenomenological Studies
Contrary to the assumptions of many who are new to qualitative research, some
approaches such as phenomenology are highly structured and follow a systematic process
of data analysis and write up. The dissemination of findings associated with a
phenomenological study represents a type of rigor normally associated with more
traditional forms of reporting. However, it is important to note that the overarching tenets
associated with qualitative writing, such as capturing participant voices, writing with the
audience in mind, and researcher reflexivity continue to hold true.
According to Moustakas (1994), individuals are surprised at the level of structure in
phenomenological research despite the highly sensitive topics that are often studied (e.g.,
childhood abuse, insomnia, etc.). The reader should come away with a highly empathetic
response leading to a deeper understanding of the participants experience as a result of
reading a phenomenological narrative. To accomplish this effect a complete and accurate
description of the experience is needed. Consequently, the data analysis and organization
of the report are key activities. Moustakas suggests a rhetorical structure in creating this
research manuscript that looks strikingly similar to traditional reports, complete with
separate chapters to organize the report.
The first chapter normally contains an introduction, statement of topic and outline and
includes research questions, purpose of the study, and the like. Next, a thorough review
of the literature is reported followed by a chapter devoted to the conceptual framework
driving the research and subsequent analysis. From there, a detailed discussion of
research methodology is presented, followed by the presentation of data and its analysis.
The final chapter provides a summary accompanied by implications, and outcomes.
It is common for phenomenological writers to focus more on these overall writing
structures; however Creswell (2007, pp. 184, 188-189) describes embedded rhetorical
structures that are still valuable in that they allow for the humanity of both the participant

57

and the researcher to be included in the narrative without sacrificing rigor. These are
summarized below:

The primary goal of the phenomenological writer is to effectively relay the

essence of the experience for the participants.

Authors often use a figure or table within the narrative or end with a creative
closing to provide an in-depth, exhaustive description of the phenomenon.

As a means of educating the reader, a philosophical discussion about

phenomenology and its assumptions may also be included.

Although the focus is clearly on describing the phenomenon from the vantage
point of the study participant, you are encouraged to refrain from bracketing
yourself out of the narrative in order to introduce the reflexivity you bring to the
study. This may involve an introduction to the phenomenon within an
autobiographical context (e.g. this authors reflections on first learning of the
Engineers Creed as an undergraduate engineering student would likely inform a
study on a students first engineering service-learning experience).

Basic Interpretive Studies

As stated in Chapter 2, basic interpretive research is concerned with a general
understanding of the human experience and how we make sense of these experiences
(Merriam, 2002). Because this approach lacks the added, specified purpose of a grounded
theory design, for example (e.g., the added purpose of theory development), it provides
the most flexible approach to writing about ones findings. Consequently, you are free to
use all structural and/or embedded devices to attain this understanding. However, it is
important to note that your voice as the researcher has a more prominent role in the text
as the reader draws meaning through your interpretive lens. As a result, a critical
examination of your own biases, perceptions, and assumptions prior to engaging the
participants in the study is useful.

Summary
The writing style of qualitative research reports is as diverse as the researchers who do it
and the approaches they take. It is a style that is consistent with the philosophical
foundations upon which qualitative research rests. Rather than employing passive voice
and presenting the results as removed from the researcher, qualitative studies bring the
voice of the researcher to the table. Rather than abstracting out only the targeted
information about the participants, qualitative studies present context and personal
perspective. They strive to preserve the participants own voices and represent their
experiences with thick, rich, detail. This chapter has talked about a variety of different
structural and rhetorical approaches to writing qualitative studies. In the end, however, it
is the skill and creativity of the researcher that determines the effectiveness of a study.

58

The goal of this chapter was to introduce the overall structural and embedded writing
techniques associated with several qualitative research designs. These techniques should
be seen, not as a substitute for technical or quantitative research reporting, but as a valued
tool in your quest to make the results of your educational studies tangible.

References Cited in This Section

Charmaz, K. (2006). Constructing grounded theory: A practical guide through
qualitative analysis. Thousand Oaks, CA: Sage.
Clandinin, D. J., & Connelly, F. M. (2000). Narrative inquiry: Experience and story in
qualitative research. San Francisco, CA: Jossey-Bass.
Corbin, J. M., & A. C. (2007). Basics of qualitative research: Techniques and
procedures for grounded theory. (3rd ed.). Thousand Oaks, CA: Sage.
Creswell, J. W. (2007). Qualitative inquiry and research design: Choosing among the
five approaches (2nd ed.). Thousand Oaks, CA: Sage.
Gilgun, J. F. (2005). Grab and good science: Writing up the results of qualitative
research. Qualitative Health Research, 15, 256-262.
Lichtman, M. (2006).Qualitative research in education: A users guide. Thousand Oaks,
CA: Sage.
Merriam, S. B. (2002). Qualitative research in practice: Examples for discussion and
analysis. San Francisco, CA: Jossey-Bass.
Merriam, S. B. (1998). Qualitative research and case study applications in education.
(2nd ed.). San Francisco, CA: Jossey-Bass.
Moustakas, C. (1994). Phenomenological research methods. Thousand Oaks, CA: Sage
.
National Society of Professional Engineers (2008). http://www.nspe.org/ethics/ehcred.asp
Richardson, L. (1990). Writing strategies: Reaching diverse audiences. Newbury Park,
CA: Sage.
Richardson, L., & St. Pierre, E. A. (2005). Writing: A method of inquiry. In N. K. Denzin
& Y. S. Lincoln (Eds.). The Sage handbook of qualitative research (3rd ed.). (pp.
959-978). Thousand Oaks, CA: Sage.
Rossman, G. B., & Rallis, S. F. (2003). Learning in the Field: An introduction to
qualitative research (2nd ed.). Thousand Oaks, CA: Sage.

59

Wolcott, H. F. (1994). Transforming qualitative data: Description, analysis, and

interpretation. Thousand Oaks, CA: Sage.
Yin, R. K. (2003). Case study research: Design and method (2nd ed.). Thousand Oaks,
CA: Sage.

60

8. Resources

The literature on conducting qualitative research studies in education has exploded over
the last thirty years. In addition to general overviews, books on specific approaches to
qualitative research as well as topics within qualitative research have been issued and
periodically revised. Journals and online forums have also appeared during these years of
growth. This section will describe some of the most accessible sources for those new to
qualitative research. Consulting the online catalog of Sage Publications
(http://www.sagepub.com/), the major source of educational methods texts, or other
publishers such as Jossey-Bass (http://www.josseybass.com/WileyCDA/), will yield
additional titles.

General Overviews
The major comprehensive volume on cutting-edge thinking about qualitative research is
The Sage Handbook of Qualitative Research, edited by N. K. Denzin and Y. S. Lincoln,
now in its third edition (2005, Thousand Oaks, CA: Sage). Although some of the chapters
in this volume and its previous editions are accessible to novice researchers, in general it
is a volume for more experienced researchers and scholars of qualitative methods.
Another more advanced text that focuses specifically on epistemological perspectives is
Foundations of Social Research by Michael Crotty (2003, Thousand Oaks, CA: Sage).
This text would be appropriate for researchers who want a more in-depth coverage of
epistemological perspectives beyond what is described in introductory texts.
A selection of introductory texts that are comprehensive in their coverage and helpful to
the new qualitative researcher is listed below. These volumes vary in the amount of
attention they devote to particular topics and the range of examples and aids included, but
all are readable and useful.

Bogdan, R.C., & Biklen, S. K. (2006). Qualitative research in education. (5th ed.).
Boston: Allyn & Bacon. As its multiple editions attest, this text has stood the test
of time. It clearly explains the background assumptions of qualitative research and
the fundamentals of data collection strategies. It is particularly strong in the area
of document/artifact analysis, especially photography. The data analysis and
writing sections are also helpful. Its treatment on different types of studies and
design is brief.
Creswell, J. W. (2007). Qualitative inquiry & research design: Choosing among five
approaches (2nd ed.). Thousand Oaks, CA: Sage. This volume is a new edition of
a favorite of graduate students. It describes, more clearly and extensively than
other introductory volumes, the differences between five selected approaches to
61

inquiry: narrative, ethnography, phenomenology, grounded theory, and case

study. It includes such helpful devices as fill-in-the blank problem statements and
examples of each type of study in appendices. Its treatment of data collection
methods is rather limited and it does not include critical studies as one of the
selected approaches.
Jones, S. R., Torres, V., & Arminio, J. (2006). Negotiating the complexities of qualitative
research in higher education: Fundamental elements and issues. New York:
Routledge. This volume is unique in that it is situated specifically within a higher
education context. It covers all of the basic issues relating to design, data
collection, analysis, and writing, with an interest in intentional practice. It is
particularly strong in its consideration of social identities.
Mason, J. (2002). Qualitative researching (2nd ed.). Thousand Oaks, CA: Sage. This
compact guide from a British scholar is organized around decisions that the
researcher must make. It treats all of the basic design, data collection, and analysis
areas, but does not focus on the more philosophical aspects of qualitative
research.
Merriam, S. B. (1998). Qualitative research and case study approaches in research (2nd
ed.). San Francisco: Jossey-Bass. Expanded from the first edition, which focused
only on case studies, this volume has been a favorite of many newcomers to
qualitative research. It is especially good on data collection techniques, but also
has sections on data analysis, trustworthiness, ethics, and writing.
Patton, M. Q. (2002). Qualitative research & evaluation methods (3rd ed.). Thousand
Oaks, CA: Sage. This comprehensive volume devotes extensive attention to the
philosophical underpinnings and theory bases supporting qualitative research, as
well as design, data collection, and analysis activities. It has extremely helpful
charts on such topics as sampling approaches and observation choices. Examples
of field notes, interview questionnaires, and analysis tasks are included. The
authors extensive experience in evaluation research leads to the inclusion of
substantial material on evaluation. He illustrates with parables, cartoons, and
personal experiences, which some readers find interesting and others find
excessive.
Punch, K. F. (1998) Introduction to social research. Thousand Oaks, CA: Sage. This
book is a general overview of social research, covering quantitative, qualitative,
and mixed methods approaches. Qualitative approaches are discussed in three
chapters, which cover research design, data collection, and data analysis. The
author mixes practical advice with philosophical aspects in a way that is more
accessible to novice researchers than some of the more advanced texts. It also
provides numerous examples through summaries of published studies.
Rossman, G. B., & Rallis, S. F. (2003). Learning in the field: An introduction to
qualitative research (2nd ed.). Thousand Oaks, CA: Sage. This book takes three

62

student researchers through their studies and thus provides practical insights on
the basics of how to approach the research act, data collection, analysis, and
reporting the study. It provides examples of the data collected from these three
hypothetical studies. Some newcomers find the concrete approach especially
helpful; others prefer less narrative.

Helpful References on Specific Topics

In addition to the number of general books available, a short list of volumes that treat
specific types of approaches or topics within qualitative research is below.
Approaches
For more in-depth description of the major approaches to qualitative design, the
following can be consulted.
Charmaz, K. (2006). Constructing grounded theory. Thousand Oaks, CA: Sage.
Corbin, J. M., & Strauss, A. C. (2007). Basics of qualitative research: Techniques
and procedures for developing grounded theory (3rd ed.). Thousand Oaks,
CA: Sage.
Moustakas, C. (1994). Phenomenological research methods. Thousand Oaks,
CA: Sage.
Willis, J. (2007). Foundations of qualitative research: Interpretive and critical
approaches. Thousand Oaks, CA: Sage.
Yin, R. (2002). Case study research: Design and methods. Thousand Oaks, CA:
Sage.
Design
Although many volumes treat design issues, the following are expressly devoted to this
topic. The first offers ideas on theoretical frameworks specifically.
Anfara, Jr., V. A., & Mertz, N. T. (Eds.). (2006). Theoretical frameworks in
qualitative research. Thousand Oaks, CA: Sage.
Marshall, C., & Rossman, G. B. (2006). Designing qualitative research (4th ed.).
Thousand Oaks, CA: Sage.
Schram, T. H. (2006). Conceptualizing and proposing qualitative research (2nd
ed.). Upper Saddle River, NJ: Pearson.

63

Interviewing
A number of guides to qualitative interviewing have appeared in recent years. These
describe methods for individual as well as focus group interviews. Among the most
popular are the following:
Kvale, S. (1996). InterViews: An introduction to qualitative research
interviewing. Thousand Oaks, CA: Sage. (Second edition due in 2008).
Rubin, H. (2004). Qualitative interviewing: The art of hearing data (2nd ed.). Thousand
Oaks, CA: Sage.
Seidman, I. (2005). Interviewing as qualitative research (3rd ed.). New York:
Teachers College Press.
Stewart, D.W, Shamdasani, P. N., & Rook, D. W. (2006). Focus groups (2nd ed.).
Thousand Oaks, CA: Sage.
Data Analysis
For special advice on general approaches to data analysis, the following are leading
guides that provide ideas on data displays, strategies, and theoretical considerations.
Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An
expanded sourcebook (2nd ed.). Thousand Oaks, CA: Sage.
Richards, L. (2005). Handling qualitative data: A practical guide. Thousand Oaks,
CA: Sage.
Silverman, D. (2001). Interpreting qualitative data: Methods for analyzing talk,
text, and interaction (2nd ed.). Thousand Oaks, CA: Sage.
Writing
Special strategies for reporting the results of qualitative studies are described in the
following:
Holliday, A. (2007). Doing and writing qualitative research (2nd ed.). Thousand
Oaks, CA: Sage.
Wolcott, H. F. (2001). Writing up qualitative research (2nd ed). Thousand Oaks,
CA: Sage.

Dictionaries
Helpful guides for defining the specialized terms that are used in qualitative research
have become very popular among novices and experienced researchers alike. The two
below are commonly accessed.

64

Bloor, M. (2006). Keywords in qualitative methods. Thousand Oaks, CA: Sage.

Schwandt, T. A. (2007). The Sage dictionary of qualitative inquiry (3rd ed.)
Thousand Oaks, CA: Sage.

Examples
Often, an example can be one of the most illustrative means for understanding how
qualitative studies are designed and conducted. The anthology by Merriam in the
following list not only showcases examples of leading types of studies, but provides
introductory commentary to each type and author comments on each piece. Yins
anthology concentrates on examples of case studies from different disciplinary
perspectives. Creswells volume, in the general section above, presents five examples of
different approaches.
Merriam, S. (Ed.) (2002). Qualitative research in practice: Examples for
discussion and analysis. San Francisco: Jossey-Bass.
Yin, R. (2004). The case study anthology. Thousand Oaks, CA: Sage.
Engineering education journals are beginning to publish some qualitative studies as well.
The best source is the Annals of Research on Engineering Education
(http://www.areeonline.org/). Some pieces are also being published in the Journal of
Engineering Education (http://www.asee.org/publications/jee/).

Qualitative Research Journals

A variety of journals on various kinds of qualitative research approaches have been
established in recent years. The list below contains four of the leading general titles, two
in print format and two online.
International Journal of Qualitative Studies in Education
http://www.tandf.co.uk/journals/tf/09518398.html
Qualitative Inquiry
http://www.sagepub.com/journalsProdDesc.nav?prodId=Journal200797
The Qualitative Report (online-Nova Southeastern University)
http://www.nova.edu/ssss/QR/index.html
Qualitative Research Journal (online-Association for Qualitative Research)
http://www.latrobe.edu.au/aqr/index.php?option=content&task=view&id=17&Ite
mid=35
65